HAL Université des Antilles
Not a member yet
25497 research outputs found
Sort by
Déterminants environnementaux de la distribution spatiale des arbres au cours des stades ontogénétiques en forêt tropicale
No full textAlthough it plays an important role in the functioning of tropical forests and the ecosystem services it provides, the understorey is much less studied than canopy trees, and therefore less well taken into account in models predicting the fate of these forests in the face of anthropogenic changes. The thesis proposes to address this need for knowledge by studying the processes governing the spatial distribution of trees during their ontogeny, using innovative approaches and new data based on LiDAR technology, field inventories of understorey trees and innovative statistical modelling approaches.Les études écologiques sur les forêts tropicales se concentrent principalement sur la canopée, laissant le sous-bois, zone de recrutement des arbres, encore peu étudié. Cette thèse propose d’adresser ce besoin de connaissances par l’étude des processus régissant la distribution spatiale des arbres au cours de leur ontogénie. Le Chapitre I évalue la capacité de différentes méthodes d'acquisition LiDAR à estimer les conditions micro-environnementales lumineuses dans le sous-bois d'une forêt tropicale dense. Le Chapitre II évalue l'importance des ressources en lumière et en eau sur la répartition spatiale des espèces d'arbres en Guyane française, et le Chapitre III étudie les variations de l'effet de la ressource lumineuse sur la répartition spatiale des espèces au cours de leur ontogénie
Molecular characterisation of the invasive terrestrial nemertean Geonemertes pelaensis: long and complex mitogenome and presence of NUMTs
No full textInternational audienceThe complete mitochondrial genome of the invasive terrestrial nemertean Geonemertes pelaensis Semper, 1863 (Nemertea: Prosorhochmidae) was sequenced from two specimens collected in geographically distant French overseas territories—Martinique in the Caribbean and New Caledonia in the South-West Pacific. In both specimens, the mitogenome contained 13 protein-coding genes, two rRNA genes, and 21 tRNA genes, and was unusually large, approaching 32 kb. The two genomes differed by only four single nucleotide polymorphisms and one indel. A comparison with 22 cox1 sequences available in GenBank confirmed this high level of genetic conservation, suggesting a recent introduction from related source populations. The extraordinary length of the mitogenome was largely attributable to two extended regions comprising only tRNA genes and long intergenic sequences. These results were contrasted with data from an unpublished SRA sequencing project (SRS20559370) of an unlocalized specimen identified as G. pelaensis; its reconstructed mitogenome was only 18 kb in length (14 kb shorter) and showed extensive sequence divergence. Phylogenetic analyses placed this specimen as the sister lineage to G. pelaensis, highlighting the need for further investigation of this taxon. In the Martinique specimen, several NUMTs (nuclear mitochondrial pseudogenes) were also detected, which could complicate future studies relying solely on Sanger sequencing. Sequencing additionally revealed prey DNA from the gut contents of both worms: the New Caledonian specimen had consumed an unidentified noctuid moth, while the Martinique specimen had likely fed on the invasive cockroach Periplaneta australasiae (Fabricius, 1775), itself an introduced species
Projet PUMAT -Pour Un Maraîchage Attractif en Martinique et en Guadeloupe : bilan et perspectives
No full textAccompagner l’évolution des systèmes agricoles vers une moindre dépendance aux produits phytopharmaceutiques en se concentrant sur les méthodes préventives et sur les évaluations élargies des pratiques de protection des cultures et en les appréciant à travers leur durabilité. Dans ce numéro, nous abordons la conception d’outils et de stratégies de prévention, le développement de méthodes de lutte non chimique contre différents ravageurs et pathogènes, ainsi que l’évaluation, la définition d’indicateurs, la diffusion et la structuration de pratiques, depuis l’échelle de la parcelle jusqu’à celle des territoires et des filières. Ces projets ont été financés dans le cadre des appels à projets Ecophyto Recherche & Innovation lancés en 2019 : - Les approches globales pour limiter l’utilisation des produits phytopharmaceutiques - Durabilité des systèmes de productions agricoles alternatifs évitant ou limitant l’utilisation des produits phytopharmaceutiques"National audienceThe ECOPHYTO PUMAT project identified several levers that promote the adoption of agroecological practices at productive scales (farm size, diversification, training), commercial scales (certification), and collective scales (producer organizations). Labour, up to 90% of total costs—and pricing are central factors. Consumers are willing to pay a premium for pesticide-free products but do not attach importance to the marketing channel. The project’s systemic approach demonstrates that preventive innovationconstitutes a collective investment, which requires trust and transparency, and that the tension between alliances and competitiveness questions the coherence of transition pathways.Le projet ECOPHYTO PUMAT a identifié plusieurs leviers favorisant l'usage de pratiques agroécologiques aux échelles productives (surface, diversification culturale, formation), commerciales (certification) et collectives (groupements de producteurs). Le travail, représentant jusqu'à 90 % des coûts, et les prix, sont centraux. Les consommateurs sont prêts à payer plus pour du sans pesticides mais sans attacher d'importance au mode de commercialisation. La démarche systémique du projet montre que l'innovation préventive est un investissement collectif, qui exige confiance et transparence, et que la tension entre collaboration et compétitivité interroge la cohérence des transitions
Dissolved organic matter dynamics and chemistry under fungal activity: A microcosm incubation with litter differentially 13C-labelled
No full textInternational audienc
Species introduction woes: How human‐mediated crop transport can influence community dynamics through mutualist displacement
No full textInternational audienceAgriculturally relevant crop species are universally associated with complex communities consisting of antagonists, commensals and mutualists. When the domesticated species are introduced into new habitats, some community associates can be left behind, and the consequences of such losses can be difficult to study. The cultivated fig Ficus carica and its community of mutualists and antagonists (wasps, nematodes and fungi) provide an excellent system to examine community shifts in this context. Cultivated F. carica can suffer significant losses due to a disease caused by Fusarium fungi. Through demographic surveys of F. carica figs in the USA and their wild European relatives as well as phylogenomic analysis of Fusarium species associated with fig and crop hosts, our research examines how species introductions can influence disease prevalence and severity. We found significant negative correlations between the presence of nematodes and fungi in wild F. carica figs in Europe. We observed similar negative correlations across wild Neotropical figs despite profound environmental differences. These observations suggest that nematodes are key components of fig communities and may directly mitigate or out‐compete pathogenic fungus. Following introduction from its endemic Mediterranean habitat to North America, F. carica crops lost an association with one endemic nematode. Consequently, we found F. carica figs in the USA to have significantly more fungal infestation than their European counterparts. Furthermore, transplantation through North America has introduced multiple novel fungal species into the F. carica system, which likely causes more severe disease for the US fig industry. Synthesis and applications : Similar negative effects associated with community disruption may be a common consequence of the introduction of many domesticated species. While it is difficult in many biological communities to successfully identify important associates and the effects of their potential losses during introduction, the biology of the fig community allows us to examine this context with greater clarity. We argue that fungal diseases in F. carica in the USA could be managed with the re‐introduction of nematodes. Further, nearly all crop species suffer losses due to a Fusarium pathogen, and nematodes may represent an unexpected and cost‐effective method of disease management beyond figs
Fine-Grained Assignment of Unknown Marine eDNA Sequences Using Neural Networks
No full textInternational audienceEnvironmental DNA (eDNA) metabarcoding is an innovative tool that is transforming ecological research. It offers a simple and effective method for simultaneously detecting numerous species across a wide range of environments. The method relies on assigning DNA sequences sampled from the environment to taxa, which is straightforward for species that have already been sequenced and are represented in reference databases. However, existing bioinformatics tools often fail to deliver accurate, fine-grained assignments when target species are absent from these databases. This limitation arises from handcrafted classification thresholds that do not account for nucleotide positional information. Here, we propose a deep neural architecture specifically designed to exploit both nucleotide identity and positional patterns in short TELEO sequences. Using an in-silico validation framework based on NCBI genbank sequences, we compare our approach with several state-of-the-art bioinformatics tools (Obitools, Kraken2, Lolo), as well as alternative sequence embedding methods, under controlled conditions. Our approach yields significantly higher classification accuracy at the genus and family levels, achieving average accuracies of 94.7% at the genus level and 86.5% at the family level, substantially outperforming the tested reference-based pipelines. The method remains robust with limited training data and shows improved performance when nucleotide positional information is preserved through sequence alignment. These results demonstrate the potential of AI-powered eDNA metabarcoding to complement existing taxonomic assignment tools, particularly in contexts where reference databases are incomplete or species-level resolution is not achievable, thereby supporting biodiversity monitoring and ecosystem management
How metabolites and metabolism in aquatic biofilms reveal ecological responses to global change and their interactions
No full textInternational audienceMicrobial communities are central to ecosystem functioning, contributing to nutrient cycling, bioremediation, and primary production. Yet, understanding in situ microbial interactions and their responses to environmental changes and chemical stressors remains a significant challenge with implications for evolution, human, animal, and environmental health, as well as biotechnology. Studies of microbial metabolites and metabolism provide valuable insights into microbial activity and microbe environment interactions. They reveal community responses, including acclimatation and adaptive strategies under stress, further linking metabolomes to biogeochemical processes and phenotypic traits in dynamic environments. Also, exometabolome analysis re-veals metabolic exchanges (e.g., cross feeding) shaping community diversity and function. However, gaps in spatiotemporal metametabolome dynamics and its ties to taxonomy hinder predictions of ecosystem structure and function, limiting actionable insights for conservation and management. Microbial biofilms in temperate climates, composed of most Life Kingdoms living in various habitats and driving ecological pro-cesses, could tackle this challenge. Recent studies use functional (photosynthesis, respiration) and structural (community composition) descriptors to assess stressor impacts on these communities but often focus narrowly on autotrophs or hetero-trophs, offering incomplete insights. These methods may lack sensitivity to detect early chemical stress or microbial interactions and overlook molecular mechanisms. Omics approaches including metabarcoding, metatranscriptomics, metaproteomics, and metametabolomics emerge as vital tools to comprehensively analyze biofilm responses, revealing taxonomic, genetic, and metabolic dynamics under environ-mental stress. Especially, untargeted metametabolomics analyzes thousands of me-tabolites from microbes environment interactions, revealing stress adaptations and bioactive chemicals, but faces technical hurdles in aquatic biofilm studies. Recent advances like mass spectrometry imaging and fluxomics improve spatial and tempo-ral metabolite tracking. Integrating these methods into a unified framework could de-code microbial acclimation and chemical interactions. Further multiomics integration could bridge gaps in understanding biofilm resilience and ecological functioning in the global change context
État des lieux de la prise en charge de la ménopause par les gynécologues en Guadeloupe et en Martinique : étude observationnelle descriptive transversale
No full textContext: In 2021, individuals aged 45 and overrepresented nearly 50% of the population in both Martinique and Guadeloupe. Menopause management therefore constitutes a significant public health issue in the French West Indies. Objectives: To assess the extent to which the overall management of menopause currently offered by gynaecologists in Guadeloupe and Martinique aligns with the 2021 clinical practice guidelines issued by CNGOF and GEMVi. Secondarily, to evaluate whether this management differs according to gynaecologists’ characteristics. Materials and methods: An anonymous online questionnaire was distributed to gynaecologists in Guadeloupe and Martinique between December 2024 and April 2025. Results: The response rate was 41%. Over half of respondents reported feeling comfortable managing menopause (56.1%). Among women without a history of hysterectomy or hormonal contraception, 87.8% of practitioners diagnosed menopause based on 12 months of amenorrhoea. Among those using hormonal contraception, only 41.5% applied this same criterion after discontinuation. Just 39.0% of respondents reported offering menopausal hormone therapy (MHT) as first-line treatment for distressing vasomotor symptoms. The genitourinary syndrome of menopause was mainly treated with local hormonal or hydrating therapies (51.2% and 34.1%, respectively). Conclusion : The study revealed heterogeneous practices, with inconsistent application of current guidelines and limited use of MHT.Contexte : En Martinique et en Guadeloupe, les personnes âgées de 45 ans et plus représentaient près de 50 % de la population en 2021. La prise en charge de la ménopause apparaît donc comme un enjeu majeur de santé publique aux Antilles. Objectifs: Évaluer l’adéquation entre la prise en charge globale de la ménopause proposée actuellement par les gynécologues en Guadeloupe et en Martinique et les recommandations pour la pratique clinique du CNGOF et du GEMVi de 2021. Secondairement, évaluer si la prise en charge diffère selon les caractéristiques des gynécologues. Matériel et méthode: Diffusion d’un questionnaire anonyme, en ligne, aux gynécologues de la Guadeloupe et de la Martinique entre le décembre 2024 et avril 2025. Résultats: le taux de participation était de 41%. Plus de la moitié des gynécologues se sentaient à l’aise avec la prise en charge de la ménopause (56.1%). 87.8% des praticiens diagnostiquaient la ménopause chez les patientes sans antécédent d’hystérectomie et sans contraception hormonale par la survenue d’une aménorrhée de 12 mois. Tandis que chez les patientes sous contraception hormonale, seuls 41.5% des praticiens la diagnostiquaient en arrêtant la contraception hormonale et en attendant la survenue d’une aménorrhée de 12 mois. Seul 39.0% des praticiens proposaient un THM en première intention pour les bouffées vasomotrices invalidantes. Le SGUM était principalement traité par traitements locaux hormonaux ou hydratant (51.2% et 34.1%). Conclusion : Une hétérogénéité des pratiques a été mise en évidence, avec une application inconstante des recommandations et un recours limité au traitement hormonal de la ménopause
Early life stages of Aquitanian pike Esox aquitanicus (Teleostei, Esocidae): first description of morphological development and trials of rearing
No full textInternational audienceAquitanian pike, Esox aquitanicus, is a new esocid species described in 2014 and endemic to Southwestern France. The four main goals of the present study were to (i) obtain the first morphological data on the early life stages, (ii) test the effects of temperature on their survival and development, (iii) compare the results with the Northern pike, and (iv) produce a few individuals to exhibit them at the Aquarium of Limoges. Eggs are yellowish, demersal, and sticky. The mean diameter was 2.73 ± 0.08 mm (n = 174). The larval size at hatching was 9.09 ± 0.24 mm (n = 14). The morphological development of 15 larvae was followed during 11 weeks as well as the snout growth in relation to eye size. It appeared that already after three weeks, slight differences between the two species were apparent, and after nine weeks, this snout/eye ratio was twice greater for Northern pike than Aquitanian pike. At the end of the yolk-feeding period, which lasted from 12 days at 18 °C to 36 days at 6 °C, the overall survival rate was over 85% and larvae were of similar size, except at 6 °C. After three months, the best results in term of survival were obtained at 12 °C (60%) and for growth at 15 °C (34.11 mm), followed at 12 °C (27.85 mm). Based on these results, we were able to rear Aquitanian pike for more than 2 years in the Aquarium of Limoges in order to promote the conservation of the species to the public
Fine-Tuning Language Models for Structured Botanical Trait Extraction and Species Comparison
No full textInternational audienceAs artificial intelligence (AI) becomes central to biodiversity research, transforming morphological descriptions of species into structured data remains a challenge, particularly in botany, where species descriptions are rich in detail but lack formal structure. Despite the progress brought by transformer architectures (Vaswani et al. 2017) and their multilingual variants such as CamemBERT (Martin et al. 2020), most large language models (LLMs) are trained on general-domain English corpora, limiting their effectiveness for non-English biodiversity applications. We present a two-stage information extraction pipeline that fine-tunes LLaMA (Large Language Model Meta AI)- based LLMs (Touvron et al. 2023) to extract botanical traits from French floristic texts and enable species-level comparison at scale. Trained on a curated corpus of legacy botanical descriptions and expert-validated French question-answer pairs, the system demonstrates how domain-specific fine-tuning can bridge the gap between narrative botanical knowledge and structured trait data (Fig. 1). At the user-interface layer, we add a third stage that automatically converts the trait-value pairs generated by the question-answering module into a structured CSV table for downstream use. Our approach unifies question generation and trait extraction within a single information extraction workflow. First, predefined botanical traits (e.g., leaf shape, flower color) are reformulated as natural-language questions tailored to each species description. Then, using these questions as prompts, the model identifies and extracts the corresponding trait values: for example, extracting "lanceolate" as the standardized leaf shape from "lanceolate leaves." This converts unstructured text into consistent character-state pairs aligned with botanical ontologies. A proof-of-concept showing the input species descriptions processed by the pipeline is shown in Fig. 2. The LLaMA model was fine-tuned via Low-Rank Adaptation (LoRA) (Hu et al. 2021) on ~17,000 manually verified French question-answer pairs derived from floristic descriptions across multiple French-language flora, including: Flore de Nouvelle-Calédonie and Flore du Cameroun . Building on FloraNER (Named Entity Recognition) (Nainia et al. 2024a) and earlier CamemBERT-based pipelines (Nainia et al. 2024b), the current system, F-LoRA-QA (Nainia et al. 2025b), substantially improves structured trait generation from domain-specific descriptions. In evaluation, F-LoRA-QA outperformed an untuned LLaMA baseline with ~4× higher BLEU score (Bilingual Evaluation Understudy), a 16% gain in semantic similarity, and exact-match accuracy rising from 2% to 24%. Expert botanists further validated trait accuracy, completeness, fluency, and terminology, supporting critiques of over-reliance on automatic metrics for domain question answering (Nainia et al. 2025a). A key application is the enrichment of descriptor-based systems like Xper³ (Kerner et al. 2025), which rely on structured trait matrices to describe and differentiate taxa. Fig. 3 and Fig. 4 illustrate trait discovery, value extraction, and a comparison matrix with evidence links, supporting descriptor systems and interoperable biodiversity databases