HAL Evry
Not a member yet
19237 research outputs found
Sort by
People, Money, and the State: Understanding the Dynamics of Labour in the Age of Inflation (3rd–4th Centuries)
No full textInternational audienceOral presentation during the first session "Free Entreprise" (26 June 2025
Nouvelles formules de synthèse de la transformée de Fourier à court terme avec une fenêtre d'analyse modifiée
No full textCommunication du Thème 6 - Oral 3 - Audio et temps-fréquence (vendredi 29 août)International audienceThe Short-Time Fourier Transform (STFT) is commonly used in applications requiring the processing of non-stationarysignals. It has inspired recent theoretical work that enables the design of its synchrosqueezed versions, improving signal readabilitywhile enhancing the separation of components and the accuracy in parameter estimation. These new operators involve STFTs usingparticular analysis windows, for which the properties are insufficiently studied. A thorough exploration of these properties allowsthe introduction of several new families of signal reconstruction formulas from its STFT, while establishing theoretical links withthe reassignment and the synchrosqueezing methods. Finally, we illustrate these theoretical results with numerical simulations.La transformée de Fourier à court terme (TFCT) est un outil couramment utilisé dans de nombreuses applications nécessitant le traitement de signaux non stationnaires. Elle a suscité de nombreux travaux théoriques récents permettant le calcul de versions synchrosqueezées, offrant une meilleure lisibilité des signaux tout en améliorant la séparation des composantes et la précision de l'estimation des paramètres. Ces nouveaux opérateurs font intervenir des TFCTs utilisant des fenêtres d'analyse particulières, dont les propriétés restent peu étudiées. Une exploration approfondie de ces propriétés nous permet d'introduire de nouvelles familles de formules de reconstruction du signal à partir de sa TFCT, tout en établissant des liens théoriques avec les méthodes de réallocation et de synchrosqueezing. Enfin, nous illustrons ces résultats théoriques par des simulations numériques
When should we expect non-decreasing returns from data in prediction tasks?
No full textThis article studies the change in the prediction accuracy of a response variable when the number of predictors increases, and all variables follow a multivariate normal distribution. Assuming that the correlations between variables are independently drawn, I show that adding variables leads to globally increasing returns to scale when the mean of the correlation distribution is zero. The speed of learning depends positively on the variance of the correlation distribution. I use simulations to study the more complex case of correlation distributions with a non-zero mean and find a pattern of decreasing returns followed by increasing returns to scale - as long as the variance of correlations is not degenerate, in which case globally decreasing returns emerge. I train a collaborative filtering algorithm using the MovieLens 1M dataset to analyze returns from adding variables in a more realistic setting and find globally increasing returns to scale across 2,000 variables. The results suggest significant scale advantages from additional variables in prediction tasks
Identifying the contribution of transposable elements to cis-regulatory elements in plants
No full textInternational audienceCis-regulatory elements (CREs) are key players of gene expression spatio-temporal regulation across development and in response to the environment. While these regions can now be molecularly depicted in plants, notably using information of chromatin accessibility and DNA methylation, characterization of stress-responsive CREs is still limited. In addition, Transposable Elements (TEs) have been proposed as a source of new CREs in plants and animals, but their relative contribution to developmental and stress-induced CREs and their contribution to the emergence of regulatory networks in plants remains to be fully elucidated.In a recent study (Fagny et al., 2021 - doi.org/10.3389/fgene.2020.606285), we used a systems biology approach to investigate the enhancer-driven regulatory network of two maize tissues at different stages: leaves at seedling stage and ear-covering bracts at flowering (husk) and made the proof of concept that such enhancer-driven networks can leverage the characterization of biologically-relevant regulatory modules in plants. Moreover, we showed that enhancers specifically active in each tissue are enriched in two different TE superfamilies, highlighting the potential role of TEs in plant tissue differentiation.In parallel, we have developed SAM-seq, a methodology allowing to simultaneously analyze chromatin accessibility and DNA methylation in plants in an unprecedented manner (Leduque et al., 2024 - doi: 10.1093/nar/gkae306). Use of this method in maize seedlings revealed new epigenomic patterns for some transposable elements, which analysis was previously hampered by Illumina-based technologies due to their repetitiveness.Following these studies, we now want to investigate the contribution of TEs to plant evolution, by leveraging CRE characterization and TE annotation for several plant species including Brachypodium distachyon
Monomeric structure of influenza A virus NEP/NS2 obtained with an artificial protein highlights conformational plasticity
No full textThe influenza virus nuclear export protein (NEP)/non-structural protein 2 (NS2) is a multifunctional protein, involved in viral ribonucleoprotein export from the nucleus, genome replication enhancement, and the adaption of avian influenza to mammals. Despite the growing attention on the importance of NEP in the influenza virus lifecycle and in interspecies transmission, the molecular details of how it performs its various roles are still not fully understood. For the purpose of assisting in structural characterization, a panel of artificial proteins (αReps) were selected against influenza virus A NEPH1N1 by phage display. When complexed with αRepE4, we were able to crystallize full-length NEPH1N1, and characterize the NEP-αRepE4 interface using an integrative native MS–XL-MS strategy, revealing a folded and monomeric NEP conformation. The N- and C-termini of NEPH1N1 pack together, with the middle linker region resembling a hinge, contrasting a previous structure where NEP is dimeric and elongated. Together these structures demonstrate the plasticity of NEP, a trait which may potentially aid NEP in binding cellular and viral partners. Using isothermal titration calorimetry (ITC) we measured a nanomolar interaction between αRepE4 and NEP. Similarly, we found that αRepE4 also binds NEP from human-isolated avian H7N9 and bovine-isolated avian H5N1 influenza viruses. Owing to their high degree of conservation, αRepE4 likely has the capacity to interact with NEP from numerous influenza A virus strains. Indeed, this, combined with the nanomolar affinity measured between NEP/αRepE4 could be explored further as a broad-range therapeutic strategy and/or a tool in cellulo to understand NEP function
From male dominated-Telecommunication, Network and Cyber jobs to more win-win jobs and companies
No full textInternational audienceIn France, tech professions, and in particular those linked to cybersecurity, do lack a pool of experts. It is well known that women desert this sector in which men prevail. Several studies have been conducted to better understand this phenomenon in the general tech field. However, only few have focused on networks and telecommunications sectors. Furthermore, studies regarding the lack of women following cybersecurity curricula or choosing jobs in this sector are sparse. The aim of this paper is to present some results of a study carried out to measure the reasons and factors responsible for the lack of women in this field in France, reviewing the situation both in the preprofessional and professional periods to better understand the issue. More specifically, we have focused on university students (in a three-year-postbaccalaureate) following Networks and Telecommunications path at the Technology University Institute (IUT - Institut Universitaire de Technologie), and alumni of this curriculum who are currently working
The diffusion and consumption of fabrics in rural Ile-de-France in the late 18th century. The purchases of the customers of Philippe Chaupellet-Chaumont, merchant clothier in Milly-en-Gâtinais (1782-1784)
No full textInternational audienc
Discrimination of Oxytocin, a Behavioral Neuropeptide Hormone, and Its Structural Variants by Nanopore
No full textInternational audienceOxytocin, a neurohormone peptide, is implicated in various physiological processes, such as functions related to social behavior, reproduction, and emotional regulation. Recently, this peptide has been identified as a potential biomarker of human psychiatric disorders. Until now, no analytical tool has been able to routinely detect oxytocin and its variants, which can have different effects on their receptor binding and activity. This work uses an aerolysin nanopore and two mutants to detect oxytocin (OXT) and oxytocin-free acid (OXT-FA) differing by a single amine group. Several oxytocin variants were used to understand the effect of structure and conformational changes of the peptides on the electric signal. We identify two conformations in native conditions, allowing their discrimination. Nanopore mutants showed improved discrimination of these peptides. Molecular dynamics simulation allowed us to rationalize the competition between electroosmosis and electrophoresis on OXT and OXT-FA transport through aerolysin. This work demonstrates the potential for analyzing conformational biomarkers for diagnostic purposes
