HAL Evry
Not a member yet
19237 research outputs found
Sort by
A divide-and-conquer approach based on deep learning for long RNA secondary structure prediction: Focus on pseudoknots identification
No full textInternational audienceThe accurate prediction of RNA secondary structure, and pseudoknots in particular, is of great importance in understanding the functions of RNAs since they give insights into their folding in three-dimensional space. However, existing approaches often face computational challenges or lack precision when dealing with long RNA sequences and/or pseudoknots. To address this, we propose a divide-and-conquer method based on deep learning, called DivideFold, for predicting the secondary structures including pseudoknots of long RNAs. Our approach is able to scale to long RNAs by recursively partitioning sequences into smaller fragments until they can be managed by an existing model able to predict RNA secondary structure including pseudoknots. We show that our approach exhibits superior performance compared to state-of-the-art methods for pseudoknot prediction and secondary structure prediction including pseudoknots for long RNAs. The source code of DivideFold, along with all the datasets used in this study, is accessible at https://evryrna.ibisc.univ-evry.fr/evryrna/dividefold/home
Les défis de la communication RSE : étude de cas longitudinale d’une plateforme d’equity crowdfunding
No full textInternational audienc
β-Cyclodextrin-derived alternating poly(glyco-triazole)s behave as amylose mimics
No full textInternational audienceCarbohydrate-derived polymers combine attractive features like abundant renewable resources, large stereochemical diversity and defined functionalization of the carbohydrates. Starting from β-cyclodextrin, a diazido heptasaccharide was regioselectively obtained in a few steps. It was used as a prepolymer for the A2B2 synthesis of alternating poly(glyco-triazole)s by copper assisted azido alkyne cycloaddition (CuAAC) with two dialkynes of different length and polarity, namely, 1,7-octadiyne and bispropargyl-polyetileneglycol-5. The resulting polymers were completely characterized by FTIR, NMR, MALDI-TOF-MS, SEC MALS, thermal analysis (TG and DSC), and SEM. The alternating insertion of the heptasaccharide and dialkyne in linear polymeric structures was confirmed by NMR and MALDI-TOF experiments. The water-soluble poly(glyco-triazole) containing PEG units had Mn 20,640 and Mw 39,650. The thermal properties (Tg = 27–42 °C) were close to those of amylose but were influenced by the linker. Therefore, these new poly(glyco-triazole)s could be considered as polysaccharide mimics and alternatives to modified native polysaccharides or brush polymers
Differences in maternal diet fiber content influence patterns of gene expression and chromatin accessibility in fetuses and piglets
No full textInternational audienceThis study investigates the impact of maternal gestation diets with varying fiber contents on gene expression and chromatin accessibility in fetuses and piglets fed a low fiber diet post weaning. High-fiber maternal diets, enriched with sugar beet pulp or pea internal fiber, were compared to a low-fiber maternal diet to evaluate their effects on liver and muscle tissues. The findings demonstrate that maternal highfiber diets significantly alter the chromatin accessibility, predicted transcription factor activity and transcriptional landscape in both fetuses and piglets. A gene set enrichment analysis revealed overexpression of gene ontology terms related to metabolic processes and under-expression of those linked to immune responses in piglets from sows given the high-fiber diets during gestation. This suggests better metabolic health and immune tolerance of the fetus and offspring, in line with the documented epigenetic effects of short chain fatty acids on immune and metabolic pathways. A deconvolution analysis of the bulk RNA-seq data was performed using cell-type specific markers from a single cell transcriptome atlas of adult pigs. These results confirmed that the transcriptomic and chromatin accessibility data do not reflect different cell type compositions between maternal diet groups but rather phenotypic changes triggered by the critical role of maternal nutrition in shaping the epigenetic and transcriptional environment of fetus and offspring. Our findings have implications for improving animal health and productivity as well as broader implications for human health, suggesting that optimizing maternal diet with high-fiber content could enhance metabolic health and immune function in the formative years after birth and potentially to adulthood
Interspecies predictions of growth traits from quantitative transcriptome data acquired during fruit development
No full textInternational audienceLinking genotype and phenotype is a fundamental challenge in biology. In this respect, machine learning is playing a pivotal role in systems biology. As a central phenotypic trait, fruit development and its relative growth rate (RGR) result from interactions between gene regulation, metabolism and environment. In the present study, we carried out a multispecies transcriptomic analysis of nine different fruits. To illustrate fruit transcriptomes, transcripts were first compared using multivariate methods, revealing main similar profiles. They were then used as variables to predict four growth traits, i.e. RGR, developmental progress, fruit weight and protein content, using generalised linear models (GLMs) to decipher the mechanisms involving gene expression in development. The predictions were very satisfactory despite disparities when the model did not include the entire panel of fruit species. Based on orthogroups derived from BLAST and annotated consensus sequences from gene ontology (GO) terminology, variables annotated for metabolic processes, especially those involving cell wall carbohydrates and proteins, were found to be the most effective in predicting growth. In addition, predictions were improved for RGR when introducing a seven-day lag between transcript contents and growth traits, suggesting the necessity of considering the proteins produced to enhance phenotypic trait predictions. These original results showed that growth traits can be predicted very well with GLMs based on orthogroups from multi-species transcriptomes
Perspectives nouvelles sur John Williams (1) : Héritages et postérités
No full textThe present issue of Émergences originates from an international conference held at Université Évry Paris-Saclay in December 2022, the results of which will appear in two complementary thematic issues of this journal in the course of 2025. Reflecting the latest developments in Williamsian studies, the aim, on the one hand, is to shed light on underexplored aspects of Williams’s work whilst on the other, to reconsider some of his works and their distinctive stylistic features through renewed approaches that offer fresh perspectives. This first, multidisciplinary volume brings together contributions from both emerging and established scholars in the fields of musicology and film studies. Collectively, they examine questions relating to Williams’s engagement with the past as well as the enduring legacy of his corpus, thereby moving beyond the divisive notions of plagiarism, paraphrase, and originality (Orosz 2015). The first four contributors explore how Williams draws from diverse musical traditions in order to enrich his own compositional practice. Among others, this entails borrowings, pastiches, references to Western art music and key figures in film music (such as James Bernard or Bernard Herrmann), or even self-referential gestures toward his earlier works in a nostalgic turn. In the final four articles, the focus is instead on defining and interrogating John Williams’s legacy, particularly his aesthetic and stylistic influence on major intergalactic audiovisual epics in both American and French productions.Le présent numéro puise sa source dans le colloque international tenu à l’Université Évry Paris-Saclay en décembre 2022, qui donnera lieu à deux numéros thématiques complémentaires d’Émergences dans l’année 2025 prenant la mesure des récentes études williamsiennes. Il s’agit d’une part, d’éclairer un certain nombre de pans de la production de Williams laissés dans l’ombre dans la littérature existante, et d’autre part, de reconsidérer plusieurs œuvres du corpus ou des traits stylistiques caractéristiques, en mobilisant des approches renouvelées qui les éclairent sous un jour inédit. Le premier dossier, pluridisciplinaire, accueille les articles de chercheuses et chercheurs jeunes comme confirmés, issus de la musicologie et des études cinématographiques. Ensemble, ils explorent les questions du rapport au passé et de la postérité du corpus williamsien, dépassant les notions clivantes de plagiat, de paraphrase, et d’originalité (Orosz 2015). Via différents prismes, les quatre premiers contributeurs analysent tout d’abord comment Williams puise dans des traditions musicales variées pour enrichir sa propre pratique compositionnelle – emprunts, pastiches, clins d’œil à la production musicale « savante » et à des figures phares de la musique à l’image (comme James Bernard ou Bernard Herrmann), voire, dans un mouvement nostalgique, à sa propre production antérieure. Les quatre derniers articles proposent, quant à eux, d’interroger et de délimiter l’héritage de John Williams, son influence esthétique et stylistique dans plusieurs grandes épopées intergalactiques audiovisuelles, américaines et françaises
Usability Evaluation of Integrated and Separated Interfaces in an Immersive Authoring Tool based on Panoramic Videos
No full textInternational audienceImmersive authoring tools have emerged as key enablers for trainers-designers to create Virtual Reality Learning Systems (VRLS) without requiring extensive programming skills. However, the design of such tools presents significant challenges in terms of interaction, usability, and interface complexity. These challenges underscore the need to validate an appropriate design interface to ensure these tools can be effectively utilized by non-technical users. This study evaluates the usability of an immersive authoring tool for VRLS using interactive panoramic videos. Two types of interfaces were compared: one that integrates storyboarding and rendering visualization, and one that separates these functionalities. Quantitative and qualitative data were collected from 24 participants divided into two groups. The results indicated that the separated interface was more effective, efficient, and satisfactory, particularly for more complex scenarios. Moreover, the integrated interface led to more pronounced symptoms of cybersickness while motivation levels did not differ significantly between the two groups. These findings highlight that integrating all functionalities into a single interface may not always be the best approach, especially for complex tasks, as it can lead to decreased usability and increased physical discomfort
Bi-Objective Optimization of a Flow Shop Scheduling Problem Under Time-of-Use Tariffs
No full textInternational audienceTime-of-use (ToU) tariffs flexibly offer markedly cheap electricity prices to industrial and residential users during off-peak periods, encouraging them to shift their peak electricity demands in valley periods. Although ToU tariffs play a crucial role in balancing electricity supply and demand, especially in energy-intensive industries, the best trade-off between industrial performance and energy costs has not been well explored. Manufacturing consumes a substantial amount of energy, primarily in the form of electricity, leading to imbalances in power consumption. The flow shop scheduling (FSS) model is one of the most prevalent models in manufacturing. To explore the significant role of ToU tariffs in manufacturing, this study addresses a bi-objective FSS problem under ToU tariffs. The objective is to find the optimal balance between customer satisfaction and total electricity cost. A tight mixed integer programming model is developed to solve this NP-hard problem using business optimizers. On the bases of the problem properties demonstrated in this study, valid inequalities are designed to reduce the solution space of the problem. For small-scale instances, an improved ɛ -constraint method is presented to find the Pareto front. For medium and large-scale instances, a two-stage fruit fly optimization (TFFO) algorithm is developed to obtain the near Pareto front. Experimental results demonstrate the efficiency and effectiveness of the proposed model and algorithms. Note to Practitioners - Scheduling for complex systems remains a formidable challenge in manufacturing. Energy cost saving is a major objective for all energy-intensive industries. Effective scheduling is crucial for businesses achieving eco-friendly performance, especially under ToU tariffs. This study aims to provide efficient scheduling model and methods that can guide decision-makers in fostering ecological transitions. The ɛ-constraint method can find globally optimal solutions within given constraints. This situation is particularly beneficial for small-scale production systems requiring high accuracy. The TFFO algorithm can handle complex industrial environments and enhance production efficiency. Additionally, the TFFO algorithm is flexible and extensible, enabling it to be generalized to other production scenarios. Overall, the proposed model and algorithms lay a solid foundation for achieving efficient scheduling under ToU tariffs
Novel Lightweight Hydraulic Integration Methodology for Robotic Applications
No full textInternational audienceHydraulic integration is one of the novel technologies in the field of robotics and assistive devices. Researchers have applied recent technologies starting from non-conventional machining methodologies and ending with additive manufacturing of metals. However, those methodologies have several drawbacks related to the cost, time, and robot weight. This motivates the research of new methodologies toward developing compact, cost-effective, and lightweight hydraulic integrated robotics mechanisms. This paper presents a novel methodology for fabricating hydraulically integrated parts by using lightweight and high-strength materials. The proposed materials, fabrication steps and the obtained results are thoroughly discussed. Silicon pipes are used for building the network of internal passages. This network is built inside a 3D-printed mould which is designed accordingly. A theoretical study is conducted for the newly manufactured hydraulic parts to define the design parameters and the working pressure at which the manufactured parts can withstand in addition to stresses and deformations analysis. The theoretical results are validated using finite element modelling (FEM) and experimental testing. The simulation results were consistent with the FEM results and the experimental results by 95% and 90% respectively. The proposed methodology is cheap and simple to implement in fabrication. Hence, this methodology can be used to fabricate hydraulic integrated components in hydraulically actuated humanoid robots successfully
What do my users want? Leveraging users insights to improve recommender systems in eWOM communities
No full textInternational audienceeWOM (electronic word-of-mouth) communities not only help their users to gain insights through the exchange of information about products, but also to make the right purchase decisions on the basis of other users' opinions. The vast number of reviews and ratings contain plenty of useful information and recommender systems are an effective tool for filtering them and providing users with the information they are looking for. However, traditional recommender systems use the rating as an input to recommend items, which leads to the cold-start problem and data sparsity. The aim of this paper is to reduce the undesirable outcomes caused by these problems and to optimize the predictive outcomes of the recommendations in the context of eWOM communities. To this end, we propose a hybrid recommender system that combines Social and eWOM variables as an input and uses the Kmeans algorithm for dimensionality reduction and the collaborative filtering SVD++ algorithm to optimize the accuracy of recommendations. Our results show that recommender systems based on users' behavioral data from eWOM communities improve recommendations compared to other recommender systems that use different variables as an input and PCA as a dimensionality reduction technique