En savoir plus

Pauline Maisonnasse
M2, VIM
Interaction entre virus de la grippe (H5N1 et H1N1) et cellules dendritiques du poumon chez le porc.
C'est un modèle intéressant puisque c'est chez cet animal que le virus peut être modifié et devenir potentiellement dangereux pour l'homme, mais en plus, sa peau et ses poumons sont très semblables à ceux de l'homme. Jusqu'à présent, j'ai principalement phénotypé les cellules dendritiques présentes dans le poumon porcin (cela avait été fait pour la peau uniquement, ou bien pour le poumon murin). Je commence en ce moment les infections de ces différentes cellules dendritiques pour déterminer le rôle de chaque sous-population dans la réponse immunitaire.
Zorana Ratkovic
Thèse 1A, MIG
Information Extraction (IE) involves automatic methods of extracting and structuring knowledge from free text.
I work in the field of Natural Language Processing, and I am looking at how semantic and syntactic analysis of text can help improve IE in the biology domain.
Florent Laferrière
Thèse 3A, VIM
Characterization of prion infectious particles by ultracentrifugation techniques: link with infectivity, neurotoxicity and seeding potential.
Je sépare les agrégats de cette protéine pathologique présents dans un cerveau infecté, selon des critères de taille, forme, densité, ce par des techniques de gradients en ultracentrifugation, et j'analyse ces populations d'agrégats par des techniques de biochimie (caractéristiques physico chimiques) , de bioessai et test d'infection en cellule (pour l'infectiosité et la neurotox) et le potentiel d'amplification par des techniques de réplication in vitro type "PCR de Prion".
Edin Hamzic
Thèse 1A, GABI-GIS
Innate and adaptive immune parameters controling response to coccidiosis in chicken
The goal of the PhD project is to decipher the genetic parameters underlying the immune response (innate and acquired) of chickens to coccidiosis (infection by Eimeria) and how this can be applied to chicken selection. The project will be focused on the study of interaction between Cobb-Vantress broilers and the pathogen Eimeria maxima, and will benefit of the knowledge acquired in the study of experimental populations infected by Eimeria tenella. The project will combine detailed analysis of specific target genomic regions containing genes that are coding for key molecules in innate (MBL) or acquired (MHC) immunity, and pangenomic approach using high-density SNP chip to study association of genetic variation to the observed phenotypes.
Rim Touhami
Thèse, AgroParisTech
Conception and evolution of ontologies for n-ary relation annotation in scientific data tables
The general framework of this research is the representation and engineering of knowledge. This research concerns more precisely the question of the construction and the evolution of an ontology in a data integration system which allows local bases to be enriched with data extracted from the Web. The ontology is central in the data integration system and allows (i) the indexation of data in the local bases, (ii) the extraction and the semantic annotation of Web data with n-ary relations defined in the ontology and (iii) the flexible and unified querying of the local data and the data extracted from the Web.
The quality of the semantic annotation and the flexible querying relies on the domain ontology. In order to improve the approach and ensure its perenniality, the object of my PhD thesis is to:
1- Propose a relevant model for an Ontological and Terminological Resource (OTR) dedicated to the task of n-ary relations annotation in scientific data tables (which are extracted from Web documents).
2- Propose a method for the management of the OTR evolution which should be able to take into account different types of changes: (i) changes explicitly required by ontologists, (ii) changes due to an alignment with external ontologies, (iii) changes required after an evolution to preserve the OTR quality assurance criteria (iv) changes required after an evolution to preserve the annotation bases quality.

Angella Nzonza
Thèse 2A, VIM
Development of a new vaccinal approach against a zoonotic virus : The West Nile Virus
West Nile Virus (WNV) is a zoonotic mosquito-transmitted flavivirus that can infect and cause disease in mammals including humans. There is no specific treatment or vaccines for WNV in humans. Our project aims at developing a WNV vectored vaccine based on a new approach which consists in using a fish rhabdovirus: Viral Hemorrhagic Septicemia virus (VHSV). Its genome consists of a single-stranded negative-sense RNA molecule of approximately 12 kb encoding five structural proteins N, P, M, G and L and, unlike the other rhabdovirus, an additional NV non structural protein. The reverse genetics system developed for VHSV offers the possibility of removing or adding genes in the viral genome and generating recombinant viruses (rVHSV). This recombinant virus can be considered as a flexible vector since there are no structural constraints (the size of the viral particle increases with the size of the genome, thus extra genes can be easily introduced). Moreover this recombinant virus is naturally inactivated at temperatures above 20°C, rendering it safe for warm-blood vertebrates. In this study, we generated the recombinant virus rVHSV-SPG-EWNV, expressing at its surface the envelope E glycoprotein (EWNV), the major antigen involved in host response against WNV, and used its as an inactivated vaccine. The EWNV gene fused to the G VHSV signal peptide (SPG), was inserted as an additional gene between the N and P cistrons. Preliminary data on the expression of EWNV at the viral membrane and protection against a WNV lethal challenge will be presented. The results suggest that this recombinant virus is able to induce a production of neutralizing antibody.
Ho Ngoc Phuong
Thèse 1A, MoSAR
Modelling of nutrients partitioning in dairy cows
Decreasing the environmental impact of dairy cow production implies an improvement in the overall efficiency of this production. Genetic selection has been the most effective way to improve milk production and subsequently the efficiencies (kg milk/kg dry matter intake) of dairy cows. Unfortunately, this selection for milk yield alone has been associated with decreased longevity of dairy cows probably via increased fertility and health problems. Central to this issue is a biological understanding of the way in which dairy cows partition their nutritional resources between production and other life functions including growth, maintenance, disease resistance, pregnancy and body reserves. This partition is not fixed across time but changes according to the physiological stage of the animal, e.g. pregnancy vs lactation, and the relative priority of each particular life function. These priorities are affected by genetic selection, indeed they are emergent properties of gene expression patterns. Finally, these priorities could only be fulfilled in the case nutrients are fully supplied. When resources are limited, a trade-off has to be made between milk production and different life functions because not all priorities can be met.
This PhD project, therefore, aims to develop a model of nutrient partitioning that: 1) takes into account the changing priorities of the cow as she progresses through different physiological stages, 2) can incorporate genetic effects on these relative priorities, and 3) realistically model how partition is altered by nutritional availability for different genotypes (G x E interaction).