HAL Université des Antilles
Not a member yet
25497 research outputs found
Sort by
Taxonomic and functional diversity of bacteria and fungi in the rye root environment is affected by landrace genotype
No full textInternational audienceThe plant microbiome is essential for plant life, with root compartments hosting diverse bacterial and fungal communities that support plant survival and productivity. This study investigates whether rye (Secale cereale L.) landraces, which may reflect local adaptations, harbor distinct root-associated microbiomes compared to modern varieties. Using metabarcoding (bacterial 16S rRNA and fungal ITS2), we characterized microbial communities in the rhizosphere and root endosphere of three local landraces and two modern varieties cultivated in two sites of the Italian Western Alps. Root compartment and growing site emerged as the main factors shaping microbial community structure, but genotype-specific differences were also observed in endospheric communities. Local landraces formed distinct microbial clusters compared to modern varieties, with differential abundances of key taxa such as Bradyrhizobium and Mortierella. Functional predictions further indicated significant variation in several bacterial metabolic pathways, suggesting that host genotype affects microbial adaptation. These findings highlight that, beyond environmental filtering, plant genotype contributes to root microbiome assembly, offering perspectives for improving agricultural resilience through genotype-informed microbiome management
Observation-only deep learning for gappy satellite-derived ocean colour data using 4DVarNet
No full textInternational audienceMonitoring optical properties of coastal and open ocean waters is crucial to assessing the health of marine ecosystems. Deep learning offers a promising approach to address these ecosystem dynamics, especially in scenarios where gapfree ground-truth data is lacking, which poses a challenge for designing effective training frameworks. Using an advanced neural variational data assimilation scheme (called 4DVarNet), we introduce a comprehensive training framework designed to effectively train directly on gappy data sets. Using the Mediterranean Sea as a case study, our experiments not only highlight the high performance of the chosen neural network in reconstructing gap-free images from gappy datasets but also demonstrate its superior performance over state-of-the-art algorithms such as DInEOF and end-to-end neural mapping schemes based CNN or UNet architectures.</div
Le septième art au carrefour de la didactique et des savoirs traditionnels : dialogue entre Rue Cases-Nègres (Euzhan Palcy) et Seawall (Mason Richards)
No full textInternational audienc
Evolution of opsin genes in closely-related species of butterflies specialized in different microhabitats
No full textInternational audienceMultiple selective pressures can shape the evolution of color vision in animals, by acting on the co- evolution of the opsin genes. How do adaptive processes shape the duplications of opsins, the evolution of their amino acids and the modification of their patterns of expression? At large phylogenetic scales, natural selection due to the contrasted light environments has been found to have a profound impact on the evolution of the opsin gene family. However, in closely-related species, species interactions due to sexual selection or competition may also influence opsin evolution. Here, we investigate the diversification of opsin sequences and their expression in closely-related blue Morpho butterfly species, living in different microhabitats, to shed light on the effect of biotic and abiotic selective pressures shaping the evolution of their opsin gene family. First, we combined genomics, transcriptomics and immunochemistry to precisely characterize the expression and the spatial distribution of the opsin proteins found in the eyes of Morpho helenor . We found unique ommatidial types compared to other butterfly species. We then investigated the evolution of opsin genes among 18 Morpho species, found signature of positive selection on two opsin genes, and identified key co-evolving amino-acids shaping the diversification of the Morpho visual system. We showed that such opsin evolution was correlated to both light environment and wing coloration, highlighting the joint effect of several selective pressures in the evolution of those proteins. Overall, our study underlines the peculiar evolution of visual systems in closely-related species specialized in divergent microhabitats
Precipitation Regimes in Haiti: A Clustering Analysis Using Chirpsv2 Data (Period 1981-2020)
No full textThis study aims to determine rainfall trends in Haiti from 1981 to 2020 using the Climate Hazards Group InfraRed Precipitation with Station data 2.0 (CHIRPSv2) dataset. Because of the scarcity of rain gauge stations in this complex topographical region, this gridded dataset was selected to compensate for the data deficit and provide an updated analysis of the country's spatiotemporal precipitation variability. Based on Kullback-Leibler divergence and cluster analysis, Haiti can be classified into four distinct zones with seven precipitation regimes (three wet, three dry, and one transitional). Wet regimes dominate the early and late rainy seasons, whereas dry regimes prevail throughout the dry season (December to March). The intermediate regime is associated with the regression of the mid-summer drought. Rainfall years are characterized by four to seven regimes that affect inter-annual and intra-annual precipitation variability. Drought periods and major climate influences are present, including the El Niño-Southern Oscillation and North Atlantic Oscillation. Climate anomalies and composites caused by fluctuations in sea surface temperature and upper-level wind patterns contribute to the transition between dry and wet regimes. This study concludes that CHIRPSv2 data provide reliable estimates for assessing precipitation trends, facilitating water resource management, and climate change adaptation. However, some limitations remain, particularly the underestimation of precipitation in mountainous locations, emphasizing the need for further research to improve local coverage
Faire la police des carrières de fonctionnaires dans les DOM : un pouvoir d’exception du Ministère et des préfets des Outre-mer au cours des années 1960 et 1970 ?
No full textInternational audienc
L’IA en formation des enseignants : éthique, littératie et reconfigurations professionnelles dans trois contextes francophones
No full textInternational audienceCette étude exploratoire examine les représentations et anticipations des professionnels de l'éducation face à l'intelligence artificielle générative (IAG), s'inspirant d'expérimentations d'intégration IA menées à l'Université de Haute-Alsace (France), l'Université du Québec à Montréal et l'Université des Antilles. La recherche compare trois territoires francophones : France/Alsace (n = 16), Québec (n = 42) et Martinique (n = 67). La recherche mobilise un cadre théorique hybride articulant théorie de l'activité (Engeström, 2015) et littératie en IA (Dale & al., 2021). L'analyse révèle cinq représentations dominantes de l'IAG (outil technique, assistant pédagogique, intelligence non-humaine, système algorithmique, automate) et quatre fonctions d'usage prioritaires (création de contenus, assistance, optimisation temporelle, soutien organisationnel). Des variations culturelles significatives émergent : pragmatisme critique québécois, ambivalence méthodique alsacienne, utilitarisme vigilant martiniquais. Les participants anticipent une reconfiguration du rôle de l’enseignant : délégation des tâches techniques à l'IA et recentrage enseignant sur les fonctions distinctement humaines. L'étude apporte trois contributions : méthodologie interculturelle francophone, opérationnalisation théorique activité/littératie, et implications pour contextualiser la littératie IA et les systèmes d’activité d’apprentissage selon les spécificités territoriales
Gilbert MATZ (1934-2021), spécialiste de l’anatomopathologiedes insectes, herpétopathologiste et pionnier de la terrariophilieen France
No full textInternational audienc
Neuroinflammation et maladies du système nerveux central aux Antilles : contribution d’un déclencheur viral au phénotype et à la progression
No full textThe French West Indies, characterized by unique environmental and genetic factors (with 90% of the population of African ancestry), provide a particularly relevant setting for targeted studies on neurodegenerative and neuroimmune diseases. This work explores the clinical, epidemiological, immunological, and therapeutic specificities of multiple sclerosis (MS), neuromyelitis optica spectrum disorder (NMOSD), and parkinsonian syndromes in this region. It highlights the importance of adapting disease analysis based on geographic and ancestral considerations, particularly for patients of African descent, who are often underrepresented in research and face disparities in healthcare access. Our research on MS demonstrates that the early initiation of high-efficacy therapies leads to better inflammation control and slower disability progression in Afro-Caribbean patients, yielding superior outcomes compared to escalation strategies while maintaining a favorable safety profile. Our findings indicate that early administration of Mitoxantrone in the management of NMOSD, which has an exceptionally high prevalence in the region, provides efficacy and safety comparable to recent therapies that are often costly and less accessible in the neighboring islands of the French West Indies, where the healthcare system remains fragile. Immunological investigations in parkinsonian patients from two distinct geographic areas, Guadeloupe and Montreal, reveal a heightened inflammatory profile in Guadeloupe, suggesting the influence of genetic and environmental factors. We hypothesize that repeated exposure to arboviruses may play a crucial role in disease progression by inducing a state of chronic inflammation. This hypothesis underscores the need for an integrated approach combining virology, genetic, and neuroimmunology to better understand the complex interactions underlying disease onset and atypical presentations, ultimately identifying appropriate therapeutic targets.Les Antilles françaises, caractérisées par des spécificités environnementales et génétiques (90% de la population d'ascendance africaine), constituent un environnement particulièrement propice pour des études ciblées sur les maladies neurodégénératives et neuro-immunes. Ce travail explore les particularités cliniques, épidémiologiques, immunologiques et thérapeutiques de la sclérose en plaques (SEP), de la neuromyélite optique (NMOSD) et des syndromes parkinsoniens aux Antilles françaises. Il met en lumière l'importance d'adapter l’analyse de ces maladies en fonction des spécificités géographiques, notamment chez les patients d’ascendance africaine, souvent sous-représentés dans la recherche et confrontés à des inégalités d’accès aux soins. Nos recherches sur la SEP montrent que l’initiation précoce de thérapies de haute efficacité permet de mieux contrôler l’inflammation et de ralentir la progression du handicap chez les patients afro-caribéens, offrant de meilleurs résultats que les stratégies d'escalade, tout en maintenant un bon profil de sécurité. Nos travaux révèlent que l’administration précoce de la Mitoxantrone dans la prise en charge de la NMOSD, dont la prévalence est exceptionnellement élevée dans la région, offre une efficacité et une sécurité comparables à celles de thérapies récentes, souvent onéreuses et peu accessibles dans les îles voisines des Antilles françaises, où le système de santé demeure vulnérable. Nos investigations immunologiques chez des patients parkinsoniens de deux zones géographiques distinctes, la Guadeloupe et Montréal, révèlent un profil inflammatoire plus prononcé en Guadeloupe, suggérant l’impact de facteurs génétiques et environnementaux. Nous émettons l’hypothèse que l’exposition répétée aux arbovirus pourrait jouer un rôle déterminant dans l’évolution de ces pathologies, induisant un état inflammatoire chronique. Cette hypothèse souligne la nécessité d’une approche intégrée combinant virologie, génétique et neuro-immunologie, afin de mieux comprendre les interactions complexes responsables du déclenchement ou de la présentation atypique de ces maladies, et d'identifier par la suite des cibles thérapeutiques