HAL-UNICE
Not a member yet
30584 research outputs found
Sort by
Prediction of miRNA-disease Associations using an Evolutionary Tuned Latent Semantic Analysis
No full textInternational audienceMicroRNAs, small non-coding elements implied in gene regulation, are very interesting biomarkers for various diseases such as cancers. They represent potential prodigious biotechnologies for early diagnosis and gene therapies. However, experimental verification of microRNA-disease associations are time-consuming and costly, so that computational modeling is a proper solution. Previously, we designed MiRAI, a predictive method based on distributional semantics, to identify new associations between microRNA molecules and human diseases. Our preliminary results showed very good prediction scores compared to other available methods. However, MiRAI performances depend on numerous parameters that cannot be tuned manually. In this study, a parallel evolutionary algorithm is proposed for finding an optimal configuration of our predictive method. The automatically parametrized version of MiRAI achieved excellent performance. It highlighted new miRNA-disease associations, especially the potential implication of mir-188 and mir-795 in various diseases. In addition, our method allowed to detect several putative false associations contained in the reference database
Study of a combinatorial game in graphs through Linear Programming
No full textIn the Spy Game played on a graph G, a single spy travels the vertices of G at speed s,while multiple slow guards strive to have, at all times, one of them within distance d of thatspy. In order to determine the smallest number of guards necessary for this task, we analyzethe game through a Linear Programming formulation and the fractional strategies it yields forthe guards. We then show the equivalence of fractional and integral strategies in trees. Thisallows us to design a polynomial-time algorithm for computing an optimal strategy in this classof graphs. Using duality in Linear Programming, we also provide non-trivial bounds on thefractional guard-number of grids and torus which gives a lower bound for the integral guard numberin these graphs. We believe that the approach using fractional relaxation and LinearProgramming is promising to obtain new results in the field of combinatorial games
Walnut husk fly, Rhagoletis completa (Diptera: Tephritidae), invades Europe: invasion potential and control strategies
No full textRhagoletis completa Cresson (Diptera: Tephritidae) is native to North America and invaded Western Europe in the late 1980s, causing important damage to its principal host, walnut (Juglans spp.). In this review, we summarize the important elements of R. completa's biology, phytosanitary status and methods used in Europe for its control, and then present the main conclusions associated with a completed risk analysis performed in 2014 to evaluate the dispersion and establishment potential of R. completa in Europe. The walnut husk fly was initially identified in Switzerland (1988) and Italy (1991), from where it spread to at least seven additional countries: France, Spain, Germany, Austria, Croatia, Slovenia and Hungary. R. completa has not reached the limits of its potential distribution. The main dissemination pathways within Europe include: (1) natural adult dissemination; (2) adult hitchhiker behaviour; and, to a lesser extent, (3) transportation of larvae in fresh fruits. R. completa host plants are widely distributed in Europe, either as isolated wild trees or in orchards, favouring the probability of fly establishment in currently fly-free areas. In addition, the European territories where Juglans species are present share biogeographic similarities. In orchards where R. completa is present and uncontrolled, 100% of walnut trees can be infested, causing losses in walnut yields of up to 80%. The negative effect is low (< 10% yield loss) under phytosanitary control, although additional costs must also be considered to support specific monitoring for R. completa. The information presented here underlines a strong need for better walnut husk fly monitoring across European countries, as well as for increasing efforts to develop biological methods to control this emerging pest
A Semantic Modeling of Moving Objects Data to Detect the Remarkable Behavior
No full textInternational audienc
Neutralising antibodies prevent PRRS viremia rebound: evidence from a data-supported model of immune response
No full textInternational audienceUnderstanding the mechanisms determining the variability in infection dynamics between hosts or strains for a given pathogen is a key issue to better understand and control infection spread. In particular, effective and constant reduction in pathogen load is desirable over infection profiles exhibiting rebounds for the health of the infected individuals and the entire herd. In this context, PRRS virus is of a particular interest. Indeed, (i) infection profiles either with or without rebound have been reported for various viral strains and host breeds; (ii) mechanisms responsible for the emergence of rebounds are unclear; (iii) PRRS virus infections are associated with highly variable global immune responses and mechanisms responsible for the infection dynamics are still poorly understood.We aimed at identifying immune mechanisms that could explain PRRSv infection rebounds using a mathematical modelling approach of the within-host dynamics. Compared to published immunological models, our model provides both an integrative and detailed view of the immune response, representing the mechanisms at the between-cell scale. We fitted the model to a set of viremia data following an experimental challenge of 240 pigs with the same dose of a virulent PRRS virus strain resulting in both rebounder (109) and non-rebounder (131) profiles. Within a profile, experimental data exhibited a wide between-host variability in infection dynamics. Between both profiles, the variability in infection dynamics preceding the rebound (i.e. during the first 20 days of the 42-day post inoculation observation period) was similar. We compared, between rebounders and non-rebounders, the set of estimated parameter values, the resulting immune dynamics and the activation levels of the underlying immune mechanisms. The activation levels were quantified by the cumulated number of viral particles or infected cells that were created or destroyed over the infection time (i.e. the flows) by mechanisms of interest: viral replication, phagocytosis of viral particles, cell infection, viral neutralisation, cytolysis (by natural killers and cytotoxic lymphocytes) and apoptosis (by TNFα antiviral cytokine) of infected cells.Compared to non-rebounders, rebounders were characterised by a higher level of immune response activation, due to higher rates of cell infection. They also exhibited higher flows of infected cell cytolysis and apoptosis, but similar viral neutralisation flows despite higher infection and viral replication flows. This points out an inadequate production of neutralising antibodies.These results would suggest that vaccines or genetic selection promoting a strong neutralising response, ideally associated with strong antiviral and cytolytic responses, should prevent against infection with rebound
Minnie : An SDN world with few compressed forwarding rules
No full textInternational audienceSoftware Defined Networking (SDN) is gaining momentum with the support of major manufacturers. While it brings flexibility to the management of flows within the data center fabric, this flexibility comes at the cost of smaller routing table capacities. Indeed, the Ternary Content-Addressable Memory (TCAM) needed by SDN devices has smaller capacities than CAMs used in legacy hardware. In this paper, we investigate compression techniques to maximize the utility of SDN switches forwarding tables. We validate our algorithm, called MINNIE, with intensive simulations for well-known data center topologies, to study its efficiency and compression ratio for a large number of forwarding rules. Our results indicate that MINNIE scales well, being able to deal with around a million of different flows with less than 1000 forwarding entries per SDN switch, requiring negligible computation time. To assess the operational viability of MINNIE in real networks, we deployed a testbed able to emulate a k = 4 Fat-Tree data center topology. We demonstrate on the one hand, that even with a small number of clients, the limit in terms of number of rules is reached if no compression is performed, increasing the delay of new incoming flows. MINNIE, on the other hand, reduces drastically the number of rules that need to be stored, with no packet losses, nor detectable extra delays if routing lookups are done in the Application-Specific Integrated Circuits (ASICs). Hence, both simulations and experimental results suggest that MINNIE can be safely deployed in real networks, providing compression ratios between 70% and 99%
High-resolution SZ imaging of clusters of galaxies with the NIKA2 camera at the IRAM 30-m telescope
No full textProceedings of the 29th Rencontres de Blois (Particle Physics and Cosmology), May 28th - June 2nd 2017, Blois (France)International audienceThe development of precision cosmology with clusters of galaxies requires high-angular resolution Sunyaev-Zel'dovich (SZ) observations. As for now, arcmin resolution SZ observations (e.g. SPT, ACT and Planck) only allowed detailed studies of the intra cluster medium for low redshift clusters (z<0.2). With both a wide field of view (6.5 arcmin) and a high angular resolution (17.7 and 11.2 arcsec at 150 and 260 GHz), the NIKA2 camera installed at the IRAM 30-m telescope (Pico Veleta, Spain), will bring valuable information in the field of SZ imaging of clusters of galaxies. The NIKA2 SZ observation program will allow us to observe a large sample of clusters (50) at redshifts between 0.4 and 0.9. As a pilot study for NIKA2, several clusters of galaxies have been observed with the pathfinder, NIKA, at the IRAM 30-m telescope to cover the various configurations and observation conditions expected for NIKA2
Persuasive Argumentation and Emotions: An Empirical Evaluation with Users
No full textInternational audienceIn everyday life discussion, people try to persuade each other about the goodness of their viewpoint regarding a certain topic. This persuasion process is usually affected by several elements, like the ability of the speaker in formulating logical arguments, her confidence with respect to the discussed topic, and the emotional solicitation that certain arguments may cause in the audience. In this study, we compare the effect of using one of the three well-known persuasion strategies (Logos, Ethos and Pathos) in the argumenta-tion process. These strategies are used by a moderator who influences the participants during the debates. We study which persuasion strategy is the most effective, and how they vary according to two mental metrics extracted from electroencephalograms: Engagement and workload. Results show that the right hemisphere has the highest engagement when Logos arguments are proposed to participants with Neutral opinion during the debate. We show also that the Logos strategy solicits the highest mental Workload, and the Pathos strategy is the most effective to use in argumentation and to convince the participants
Wideband multiple diversity tensor array processing
No full textInternational audienceThis paper establishes a tensor model for wideband coherent array processing including multiple physical diversities. A separable coherent focusing operation is proposed as a pre-processing step in order to ensure the multilinearity of the interpolated data. We propose an ALS algorithm to process tensor data, taking into account the noise correlation structure introduced by the focusing operation. We show through computer simulations that the estimation of DoA and polarization parameters improves compared to existing narrowband tensor processing and wideband MUSIC. The performance is also compared to the Cramér-Rao bounds of the wideband tensor model
Implications of Efimov physics for the description of three and four nucleons in chiral effective field theory
No full textInternational audienceIn chiral effective field theory the leading order (LO) nucleon-nucleon potential includes two contact terms, in the two spin channels S=0,1, and the one-pion-exchange potential. When the pion degrees of freedom are integrated out, as in the pionless effective field theory, the LO potential includes two contact terms only. In the three-nucleon system, the pionless theory includes a three-nucleon contact term interaction at LO whereas the chiral effective theory does not. Accordingly arbitrary differences could be observed in the LO description of three- and four-nucleon binding energies. We analyze the two theories at LO and conclude that a three-nucleon contact term is necessary at this order in both theories. In turn this implies that subleading three-nucleon contact terms should be promoted to lower orders. Furthermore, this analysis shows that one single low-energy constant might be sufficient to explain the large values of the singlet and triplet scattering lengths