libra UniNe
Not a member yet
28871 research outputs found
Sort by
Higher spatial than seasonal beta diversity of soil protists along elevation gradients
Get PDFBiodiversity patterns along elevation gradients have long been studied for plants and animals, but only quite recently for soil microorganisms, especially protists (eukaryotes excluding plants, animals, and fungi). Microorganisms have shorter generation times than macroorganisms, and their abundance, diversity, and community structure are known to vary rapidly in response to abiotic and biotic factors. If microbial diversity varies more seasonally than spatially, a single sampling campaign along an elevation gradient, with contrasted phenologies, could introduce bias into biodiversity studies comparing multiple elevation gradients across different seasons, habitats, regions or latitudes. To address this question, we investigated the relative magnitude of spatial versus temporal diversity (alpha diversity) and community turnover (beta diversity) of soil protist communities along elevation gradients in two distant European mountain ranges. We collected soil samples in forests and grasslands below the treeline along five elevation gradients in two consecutive seasons (spring and summer) in the Spanish Sierra Nevada and the Swiss Alps, covering two distinct biogeographic regions. Using general eukaryotic primers and amplicon sequencing of soil environmental DNA, we decomposed total protist amplicon sequence variants diversity into local alpha- and beta diversity components and identified climatic and edaphic predictors of biodiversity patterns using redundancy analyses. Soil protist communities varied spatially within and among transects but temporal turnover was comparatively low. The best edaphic predictors of community variations were the same in spring and summer, but their explanatory power differed among seasons. The dominant spatial component of beta diversity suggests that patterns of soil protist communities along elevation gradients are more strongly driven by spatial heterogeneity than inter-seasonal turnover. Thus, in temperate climates, our results suggest that sampling only once between the end of spring and late summer across an elevation gradient does not introduce bias due to phenological differences when comparing beta diversity across multiple gradients.
Spatio-temporal dynamics of soil protists communities were studied in forests and grasslands below the tree line along five elevation gradients in the Spanish Sierra Nevada and the Swiss Alps during two consecutive seasons (spring and summer).
The total diversity of soil protist communities was predominantly shaped by beta-diversity components with spatial heterogeneity rather than temporal turnover as the main driver of soil protist community composition.
Community dissimilarity of soil protists did not differ in response to temporal changes between habitats (i.e., forests versus grasslands)
The significant edaphic predictors of protist community composition were highly similar in the Swiss Alps and identical in the Spanish Sierra Nevada between both seasons, but their explanatory power varied between spring and summer.
Soil protist beta diversity patterns along different elevation gradients remained constant between seasons. This suggests that, in temperate climates, sampling at one time across an elevation gradient will not bias results stemming from phenological contrasts, allowing comparison of beta diversity patterns along such gradients between regions even if sampling is not simultaneous.tru
Two-speed genomes of Epichloe fungal pathogens show contrasting signatures of selection between species and across populations
Get PDFAntagonistic selection between pathogens and their hosts can drive rapid evolutionary change and leave distinct molecular footprints of past and ongoing selection in the genomes of the interacting species. Despite an increasing availability of tools able to identify signatures of selection, the genetic mechanisms underlying coevolutionary interactions and the specific genes involved are still poorly understood, especially in heterogeneous natural environments. We searched the genomes of two species of Epichloe plant pathogen for evidence of recent selection. The Epichloe genus includes highly host-specific species that can sterilize their grass hosts. We performed selection scans using genome-wide SNP data from seven natural populations of two co-occurring Epichloe sibling species specialized on different hosts. We found evidence of recent (and ongoing) selective sweeps across the genome in both species. However, selective sweeps were more abundant in the species with a larger effective population size. Sweep regions often overlapped with highly polymorphic AT-rich regions supporting the role of these genome compartments in adaptive evolution. Although most loci under selection were specific to individual populations, we could also identify several candidate genes targeted by selection in sweep regions shared among populations. The genes encoded small secreted proteins typical of fungal effectors and cell wall-degrading enzymes. By investigating the genomic signatures of selection across multiple populations and species, this study contributes to our understanding of complex adaptive processes in natural plant pathogen systems
La Suisse romande au miroir de la littérature médiévale
No full textCe livre est le premier ouvrage entièrement consacré aux rapports de la Suisse romande médiévale et de la littérature. Écrit sous une forme attrayante et destiné à un large public, il espère intéresser aussi les spécialistes de la chose littéraire. Son auteur, qui enseigne depuis plus de vingt ans la littérature française médiévale aux Universités de Lausanne et de Neuchâtel, ne s’est en effet pas contenté de réunir des faits rebattus, mais a tenté de mettre en perspective une matière protéiforme, qui frappera le lecteur par sa richesse et sa variété, et même d’apporter au débat quelques éléments nouveaux.
Les textes littéraires écrits en Suisse romande au Moyen Âge ne sont pas légion. En leur adjoignant tous ceux qui font des allusions à nos régions et en évoquant les personnages et les légendes qui ont pu inspirer des auteurs plus récents, on obtient cependant un beau florilège qui permet, par petites étapes buissonnières, de reconstituer et d’éclairer de larges pans de la littérature du Moyen Âge. Même si le concept de « Suisse romande » est anachronique avant la fin du xve siècle, il se dégage bel et bien de cette enquête le sentiment que, de la Reine Berthe aux Réformateurs, du Chat de Lausanne à Oton de Grandson, de Chalamala à Antitus, il se joue entre Jura, Sarine et Alpes, durant l’époque médiévale, quelque chose de fondateur de notre identité romande
Explainable Machine Learning: Approximating Shapley Values for Dependent Predictors
Get PDFThesis submitted in partial fulfillment of the requirements for the degree of Master of Science in Statistics
Supervisor:
Prof. tit. Dr. Alina Matei
Université de Neuchâtel
Faculty of Science
Institute of StatisticsModern Machine Learning algorithms often outperform classical statistical methods in predictive accuracy. This comes at the expense of model interpretability. As businesses and institutions increasingly rely on Machine Learning to support and automate decision making processes to reap the benefits of more accurate predictions, explaining these model outputs becomes more important. A universally applicable approach to explaining such complex models is based on the Shapley value, a concept originating from game theory. However, its calculation is very computer-intensive, so approximations have to be used. The state-of-the-art approach, Kernel SHAP, assumes independence of the predictors, which is unrealistic in practice. Recent research has developed improvements to incorporate dependencies between predictors. After a review of the theoretical underpinnings, the original KernelSHAP method is compared with improved versions in realistic settings, using three real-world datasets. While the improved versions are found to have smaller approximation error to exact Shapley values, they are also more computationally demanding. Further improvements are discussed and possible research directions are suggested. The thesis is structured as follows: After introducing explainable machine learning in chapter 1, the Shapley value and its applications to model explainability are explored in chapter 2. Chapter 3 presents methods to approximate Shapley values as well as recent improvements to these methods, which are tested on real datasets in chapter 4. Some possible directions for future research are pointed out in chapter 5, before giving a final conclusion in chapter 6. Code for the experiments of chapter 4 is found in the appendix
Exploring the (Un)sustainability of Paris 2024 and Los Angeles 2028
No full textThis chapter deals with the sustainability assessment of the Paris and Los Angeles Games. The economic, social, and political risks inherent in hosting the Olympics are well documented. The literature features a litany of problems for host cities and societies: exorbitant costs and broken budgets, ambitious urban redevelopment agendas that harm cities, socio-spatial exclusion that further marginalizes the already-vulnerable, and lasting damage to the natural environment. To manage the decline of bidders, the International Olympic Committee (IOC) engaged in organizational reforms. The chapter shows that one of the main aims of the reforms is to improve alignment between the Games and the city’s existing development trajectory and needs. By better adapting the Games to the host city, the IOC hopes to interrupt the trend of damage, decline, and bad publicity, and thereby reawaken global interest in hosting the Olympics. Through these reforms, we hypothesize that the IOC is attempting to express a commitment to long-term planning in line with the needs of the host city. While significant progress has been made, more remains to be done if the Games are to be legitimately sustainable for all
Attribution des déductions pour les contributions d’entretien et les intérêts passifs dans un contexte international
Get PDFRésumé et analyse, Tribunal fédéral – 9C_47/2024 et 9C_48/2024, destiné à la publication, IIIe Cour de droit public, Arrêt du 23 septembre 2024 (d
Evaluation du langage oral dans l'aphasie epileptique chez l'adulte: une revue systèmatique
No full textPopulation-level transposable element expression dynamics influence trait evolution in a fungal crop pathogen
No full textThe rapid adaptive evolution of microbes is driven by strong selection pressure acting on genetic variation. How adaptive genetic variation is generated within species and how such variation influences phenotypic trait expression is often not well understood though. We focused on the recent activity of transposable elements (TEs) using deep population genomics and transcriptomics analyses of a fungal plant pathogen with a highly active content of TEs in the genome. causes one of the most damaging diseases on wheat, with recent adaptation to the host and environment being facilitated by TE-associated mutations. We obtained genomic and RNA-sequencing data from 146 isolates collected from a single wheat field. We established a genome-wide map of TE insertion polymorphisms in the population by analyzing recent TE insertions among individuals. We quantified the locus-specific transcription of individual TE copies and found considerable population variation at individual TE loci in the population. About 20% of all TE copies show transcription in the genome suggesting that genomic defenses such as repressive epigenetic marks and repeat-induced polymorphisms are at least partially ineffective at preventing the proliferation of TEs in the genome. A quarter of recent TE insertions are associated with expression variation of neighboring genes providing broad potential to influence trait expression. We indeed found that TE insertions are likely responsible for variation in virulence on the host and potentially diverse components of secondary metabolite production. Our large-scale transcriptomics study emphasizes how TE-derived polymorphisms segregate even in individual microbial populations and can broadly underpin trait variation in pathogens.IMPORTANCEPathogens can rapidly adapt to new hosts, antimicrobials, or changes in the environment. Adaptation arises often from mutations in the genome; however, how such variation is generated remains poorly understood. We investigated the most dynamic regions of the genome of a major fungal pathogen of wheat. We focused on the transcription of transposable elements. A large proportion of the transposable elements not only show signatures of potential activity but are also variable within a single population of the pathogen. We find that this variation in activity is likely influencing many important traits of the pathogen. Hence, our work provides insights into how a microbial species can adapt over the shortest time periods based on the activity of transposable elements
The Asylum Migration Crisis in The Balkans and the Road to Europe : Unfolding the Drivers Complex
Get PDFThe year 2015 saw a significant increase in the number of refugees and migrants arriving in Europe, mainly from the Middle East, Africa, and South Asia. It was so-called “European refugee (or migration) crisis”. Many of these migrants travelled via the Balkan route: After reaching Greece, they crossed the Western Balkans to reach the northern EU countries. Others took the Eastern Mediterranean route, crossing the Aegean Sea, or the Central Mediterranean route from North Africa to Italy. The geopolitical causes of this migration have been extensively analysed, with the Syrian conflict playing a central role as a push factor. The consequences and political context of the crisis—from Chancellor Merkel’s “Wir schaffen das” (“We can do it”) to its role in the UK’s Brexit from the EU—have also attracted much attention. The present paper takes a different stance, looking at the deeper structural and geographical transformations that explain why such an influx of people into Europe was possible in 2015, whereas it had never happened before in similar contexts of violence in the regions of origin. I draw on recent theoretical advances in migration theories that bridge the gap between forced and voluntary migration (push-pull-plus models, aspirations and (cap)abilities theories, etc.). I propose three long-term factors that have facilitated access to European borders but led to urgent and often dangerous migratory situations for asylum seekers: the “shortening” of distances (Europe is easier to reach), the crisis of containment policies (local help for people on the move is difficult to find), and the geographical asymmetry of rights (migrants enjoy substantial protection once in Europe, but not on route and near their country of origin). I believe that these factors played a central role in 2015 and still explain the current situation of migration to Europe and the new increase in migration observed in 2023. In this context, I interpret the current EU policy of closing borders as an attempt to (re)create a geographical buffer separating refugees from their destinations in the context of the globalisation of asylum issues.</jats:p