Chez le poulet, les réserves musculaires en glycogène disponibles au moment de la mort constituent un élément déterminant de la qualité technologique de la viande via leur effet sur le pH ultime. L’objectif de la thèse était d’identifier grâce à des approches de génomique fonctionnelle les gènes à l’origine des variations de glycogène et donc de qualité. Une approche ciblée sur l’AMP-activated protein kinase (AMPK) et plusieurs enzymes qui contrôlent le métabolisme du glycogène a suggéré que des régulations au niveau transcriptionnel d’un certain nombre de gènes (PRKAG2, GYS, PYG,…) pourraient intervenir dans le contrôle des réserves musculaires en glycogène. Cette étude, qui comparait des poulets maigre et gras a en outre révélé un niveau d’activation 3 fois supérieur chez les poulets maigres qui présentaient les réserves musculaires en glycogène les plus faibles. Des analyses transcriptomiques réalisées sur puces à oligonucléotides ont permis de mettre en évidence 439 gènes différentiellement exprimés entre individus présentant des différences de teneurs musculaires en glycogène. Parmi eux, certains comme CEBPB, PDK4 ou UGDH pourraient jouer un rôle direct dans la régulation du métabolisme du glycogène au niveau du muscle. L’analyse de données positionnelles (Quantitative Trait Loci ou QTL) a mis en évidence d’autres gènes qui de part leur régulation fonctionnelle et leur position sur le génome viennent enrichir la liste des candidats révélée au cours des études d’expression. L’ensemble de ces travaux constitue une première étape pour l’identification des gènes impliqués dans les variations de qualité de viande chez le poulet. Après validation, notamment sur des populations commerciales, nos résultats doivent à terme permettre de développer des outils d’aide à la sélection (marqueurs moléculaires) mais aussi d’optimiser les pratiques d’élevage pour améliorer la qualité des viandes de poulet. / In chickens, the muscle glycogen stores available at death are determinant for meat quality via their effect on the ultimate pH. This thesis aimed at identifying through functional genomic approaches genes causing changes in glycogen and therefore meat quality in chicken. A focused approach on the AMP-activated protein kinase (AMPK) and several enzymes that control glycogen metabolism suggested that transcriptional regulation of several genes (PRKAG2, GYS, PYG, …) could be involved in the control of muscle glycogen stores. This study, which compared lean and fat chickens also showed activation levels 3 times higher in lean chickens whose glycogen stores are the lowest. Transcriptomic analyses performed on oligonucleotide microarrays have highlighted 439 genes differentially expressed between individuals with different levels of muscle glycogen store. Among them, some as CEBPB, PDK4 or UGDH could play a direct role in the regulation of glycogen metabolism in muscle. Analysis of positional data (Quantitative Trait Loci or QTL) allowed identifying other genes which were both regulated at the functional level and located within a QTL of meat quality. Those genes enriched the list of candidates revealed by expressional studies. Altogether these studies are a first step in identifying genes involved in meat quality changes in chicken. After validation within commercial genotypes to appreciate the real interest in term of selection, our results should ultimately help to develop molecular tools useful for selection but also for optimizing rearing practices to improve the quality of poultry meat.

La connaissance des gènes exprimés dans une cellule, dans des conditions physiologiques ou pathologiques, est un élément essentiel à la compréhension des phénomènes biologiques qui la gouvernent. Parmi les technologies permettant de mesurer l'expression génique, la plus utilisée est la technologie des puces à ADN capable de mesurer l'abondance relative des gènes exprimés dans les cellules. Les puces qualifiées de pangénomiques sont supposées couvrir l'ensemble des gènes existants, soit près de trente-mille dans l'espèce humaine. La mesure, l'analyse et l'interprétation d'une telle quantité de données posent un certain nombre de problèmes et la maîtrise des méthodes d'analyse utilisées déterminera la fiabilité et la précision des informations obtenues. Le but de cette thèse est de définir des méthodes permettant de contrôler les mesures, d'améliorer l'analyse et d'approfondir l'interprétation des données transcriptomiques afin d'en optimiser l'utilisation et de pouvoir appliquer ces méthodes pour analyser le transcriptome de patient atteint de leucémie myélomonocytalre juvénile dans le but d'améliorer le diagnostic et de comprendre les mécanismes biologiques de cette maladie rare. Nous avons ainsi développé, et validé au travers de nombreux projets indépendants, un programme de contrôle qualité des puces, ainsi qu'un logiciel qui permet d'améliorer les interprétations biologiques des données microarrays basées sur les ontologies des gènes, et un outil de visualisation et d'analyse globale des voies de signalisation. Enfin, en combinant plusieurs des approches , décrites, nous avons mis au point une méthode pour obtenir des signatures biologiques fiables à des fins diagnostiques. / To understand the biological phenomena taking place in a cell under physiological or pathological conditions, it is essential to know the genes that it expresses Measuring genetic expression can be done with DNA chlp technology on which are set out thousands of probes that can measure the relative abundance of the genes expressed in the cell. The microarrays called pangenomic are supposed to cover all existing proteincoding genes, that is to say currently around thirty-thousand for human beings. The measure, analysis and interpretation of such data poses a number of problems and the analytlcal methods used will determine the reliability and accuracy of information obtained with the microarrays technology. The aim of thls thesis is to define methods to control measures, improve the analysis and deepen interpretation of microarrays to optimize their utilization in order to apply these methods in the transcriptome analysis of juvenile myelomocytic leukemia patients, to improve the diagnostic and understand the biological mechanisms behind this rare disease. We thereby developed and validated through several independent studies, a quality control program for microarrays, ace.map QC, a software that improves biological Interpretations of microarrays data based on genes ontologies and a visualization tool for global analysis of signaling pathways. Finally, combining the different approaches described, we have developed a method to obtain reliable biological signatures for diagnostic purposes.

Ophiostoma floccosum and Ophiostoma piliferum are polymorphic ascomycete fungi found throughout the world. Both species are important economically as they are known to colonise timber and cause discoloration of wood thus reducing its aesthetic value and subsequently price. Albino variants of the two species, in particular O. piliferum, are used as biological control agents to prevent sapstaining and have been used commercially for the past 15 years to reduce pitch/wood extractives in paper manufacturing. Other members of the genus include the plant pathogens O. novo-ulmi and O. clavigerum, known to have a severe effect on forest health and economy around the world. O. floccosum and O. piliferum have been demonstrated in the laboratory to be fermented in large volumes and they are particularly suitable as hosts capable of secreting extracellular recombinant proteins. This research aimed to investigate the transcriptome and molecular functioning of Ophiostoma floccosum and compare this to transcriptomic data available for Ophiostoma piliferum and other Ophiostoma species, O. novo-ulmi, O. clavigerum and O. piceae. This research contributes to the development of O. floccosum and O. piliferum as hosts for protein expression and advances the knowledge of gene expression and molecular functioning in this genus. To gain insight into the molecular functioning of O. floccosum, an expressed sequence tag (EST) collection from yeast-like growth (blastospores) was created during early phase growth. A total of 1207 EST sequences with an average length of 713 bp were identified. Clustering and assembly of the high-quality EST data set resulted in the identification of 598 unique putative transcripts (UPTs). Functional classification of these UPTs, using both homology searching and ab-initio methods, indicated that the majority of protein transcripts produced were involved in metabolism and cell proliferation. Up-regulation of mitochondrial transcripts involved in respiration and the presence of transcripts homologous to enzymes involved in the tri-carboxylic acid cycle indicated that aerobic respiration was likely the preferred method of ATP production in O. floccosum blastospores. However, the putative identification of genes encoding alcohol dehydrogenases within O. floccosum ESTs and the presence of homologues in other Ophiostoma species would suggest that these Ophiostoma species are also likely to be capable of metabolic functioning under anaerobic conditions. To identify homologous genes between Ophiostoma species, the O. floccosum EST data set was compared to 20,783 ESTs from other Ophiostoma species including O. piliferum, O. novo-ulmi, O. clavigerum and O. piceae. All UPTs identified within each of the datasets were aligned resulting in the identification of 347 clusters containing EST sequences from more than one Ophiostoma species. Six were identified that had homologues in all of the datasets excluding O. piceae. Three of the six homologous UPTs were predicted to function in core metabolism with two of the UPTs identified as encoding enzymes used in the glycolysis pathway and one encoding a 60S ribosomal protein. The other three homologous UPTs were thought to have a functional role in protein fate and were putatively identified as being a superoxide dismutase, heat-shock protein and a structural alpha-B chain tubulin gene. Of the 347 clusters, 86 of these contained transcripts identified in the O. floccosum EST datasets, and of these 86, only 10 fragments did not align with any significant homology to other fungal sequences contained in the NCBI non redundant database, indicating that the majority these transcripts are conserved in other fungal species. Predicted genes within the Ophiostoma EST datasets were also investigated to determine codon usage and to identify the presence of genes predicted to encode proteases. Both are important factors in recombinant protein expression. Protease production can severely inhibit the production of recombinant protein in fungal hosts. Based on sequence homology to known proteases, putative proteases were identified in all of the Ophiostoma species investigated with the exception of O. piceae. Homologues for all six peptidase groups were identified including a possible glutamic acid protease and proportionally high numbers of serine and metallo-protease homologues. This research constitutes the first reported findings of putative peptidases in the aspartic, cysteine, glutamic and threonine peptidase families in Ophiostoma species. Key to the over-expression of recombinant proteins is the optimisation of codons in a cloned gene to better utilise available tRNA species within the recombinant host. No codon bias was apparent between up-regulated and lower frequency transcripts in O. floccosum, O. piliferum, O. clavigerum and O. novo-ulmi. Codon usage was found to be consistent between these Ophiostoma species. However, a large difference between the codon usage in mitochondrially encoded genes compared to nuclear encoded genes in O. floccosum was indicated. To optimise the efficiency of a recombinant expression system, we sought to identify promoters in both O. floccosum and O. piliferum that may be applied to a vector system. Using EST data, the most up-regulated UPTs identified from O. floccosum and O. piliferum ESTs were a putative subunit 4 of the NADH-ubiquinone oxidoreductase protein (NADH-UR4) and a possible heat-shock protein (HSP), respectively. A unique hydrolase gene was also identified by molecular probing of O. floccosum genomic DNA. This putative 96 kd protein, called PLIP-Lg, was predicted to be a mitochondrial A1 phospholipase based on both nucleotide and predicted amino acid sequence structure and homology. These gene sequences were investigated using genome walking methods to further elucidate nucleotide sequences in the 5' and 3' directions. In silico investigation of the 5' promoter region of the genes identified a number of predicted transcription factor binding sites, including possible TATA boxes identified previously in the promoter region of an O. floccosum protein. Additionally, RT-PCR methods were used to compare the expression of these transcripts throughout growth in both the mycelial and blastospore forms. All three predicted genes were found to be transcribed throughout growth in both morphological forms and, thus, the use of their promoters in a vector system would not be limited to one morphology. However, the level of expression in blastospores compared to mycelial growth varied by up to 20 fold. Therefore, the morphological form of the fungi did influence the level of expression of these genes and is a factor for consideration for future promoter use. This PhD thesis research provides the first comprehensive investigation into gene expression and the transcriptome of O. floccosum while also providing the first comparative look into similarities between the transcriptomes of several Ophiostoma species. Subsequently, this research adds to the knowledge of metabolic functioning in Ophiostoma species and illustrates the usefulness of EST analysis in determining core molecular functioning within this group. Further to addressing these goals, the research will augment future research into various biotechnological applications for the genus, specifically the development of O. floccosum and O. piliferum as hosts for recombinant protein expression.

Inflammatory bowel disease (IBD) is a composite classification for a range of gastrointestinal disorders. The two most common forms of IBD are Crohn’s disease (CD) and ulcerative colitis (UC). The chronicity of IBD has long-term consequences on the quality of life of affected individuals. However, therapies available today are applied with limited success. It is thought that interplay between environmental triggers and commensal bacteria can cause a dysregulated inflammatory response in genetically predisposed individuals, but the exact etiology is not understood. The objective of these studies was to identify the gene expression changes which associate with differential response to therapy and with recurrence of disease following surgery. The effects of genetic variation on gene expression were also evaluated. Tissue and blood samples were collected from eligible patients. Total RNA was extracted and measured on gene expression microarrays. The raw gene expression data were analyzed in a statistical framework against variables of interest in order to assess the significance of each association. Genes were evaluated individually as well as in biological networks. A few tens of genes were identified as differentially expressed in blood between UC patients who respond to intravenous corticosteroid therapy and those who do not. Some of these genes were also shown to have high predictive value. Differentially regulated genes were also found in UC patients who experience recurrence of disease after surgery, relative to those who remain disease-free. Specifically, gene networks responsible for the regulation of cellular transport were among the major players. A comparative analysis of genotype and gene expression in the human ileum indicated that the transcription of approximately 10% of genes is influenced by variations in the genome. Results from these studies have not only contributed to our understanding of disease mechanism, but could also have medical implications. With the advance of new and less costly transcriptome technologies on the clinical stage, panels of expression markers associated with therapy response and post-operative recurrence can be used as diagnostic and prognostic tools. Data on the relationship between genotype and gene expression are already shedding new light on the function of certain genetic IBD risk variants.

Inflammatory bowel disease (IBD) is a composite classification for a range of gastrointestinal disorders. The two most common forms of IBD are Crohn’s disease (CD) and ulcerative colitis (UC). The chronicity of IBD has long-term consequences on the quality of life of affected individuals. However, therapies available today are applied with limited success. It is thought that interplay between environmental triggers and commensal bacteria can cause a dysregulated inflammatory response in genetically predisposed individuals, but the exact etiology is not understood. The objective of these studies was to identify the gene expression changes which associate with differential response to therapy and with recurrence of disease following surgery. The effects of genetic variation on gene expression were also evaluated. Tissue and blood samples were collected from eligible patients. Total RNA was extracted and measured on gene expression microarrays. The raw gene expression data were analyzed in a statistical framework against variables of interest in order to assess the significance of each association. Genes were evaluated individually as well as in biological networks. A few tens of genes were identified as differentially expressed in blood between UC patients who respond to intravenous corticosteroid therapy and those who do not. Some of these genes were also shown to have high predictive value. Differentially regulated genes were also found in UC patients who experience recurrence of disease after surgery, relative to those who remain disease-free. Specifically, gene networks responsible for the regulation of cellular transport were among the major players. A comparative analysis of genotype and gene expression in the human ileum indicated that the transcription of approximately 10% of genes is influenced by variations in the genome. Results from these studies have not only contributed to our understanding of disease mechanism, but could also have medical implications. With the advance of new and less costly transcriptome technologies on the clinical stage, panels of expression markers associated with therapy response and post-operative recurrence can be used as diagnostic and prognostic tools. Data on the relationship between genotype and gene expression are already shedding new light on the function of certain genetic IBD risk variants.

High-throughput genomics methodologies provide us the methods and solutions to the research fields of genomes, transcriptomes and proteomics. In my study, we utilized different high-throughput genomics methodologies to explore the exomes of human patients as well as transcriptomes of Drosophila melanogaster and Daphnia pulex. Exome sequencing, an efficient strategy to sequence the coding regions of the genome, helped us to identify potential disease causing mutations in two human diseases (Renal Agenesis and Cleft lip with or without palate). Although exome sequencing is mostly used to identify mutations underling Mendelian disorders, we chose informative pedigrees that were transmitted in Mendelian fashion for our study. Identification of new genes will be expected to significantly impact our understanding of human development and human diseases. RNA-Seq, a high-throughput sequencing technology to sequence cDNA in a sample, was utilized to explore the transcriptome of the microcrustacean Daphnia pulex. Daphnia pulex is a keystone species of freshwater ecosystems and are routinely used to determine the toxicity of aqueous solutions and to gauge the quality of inland water. Treatment of Daphnia pulex with different heavy metals followed by RNA-Seq analysis revealed us specific gene expression patterns that would provide new insights into their biological and toxicological responses to these environmental contaminants. Also these studies will help us to discover new genes that are not expressed during normal development. We also developed a new technology that utilizes microarray-based exome capture prior to RNA-Seq to faithfully annotate tissue specific transcriptomes in Drosophila melanogaster. We showed that this methodology was efficient for cataloging tissue-specific transcriptomes in which specific classes of genes or transcripts can be targeted for capture and sequence, thus reducing the significant sequencing depth normally required for accurate annotation.

Mon travail de thèse a porté sur l'étude des facteurs influençant le devenir de l'infection palustre. Une étude de liaison génétique basée sur des marqueurs répartis sur l'ensemble du génome nous a permis de mettre en évidence la liaison génétique de la région chromosomique 17p11-p13 avec la parasitémie. Nous avons étudié les variations génétiques des gènes HS3ST3A1 et HS3ST3B1 du chromosome 17p12 humain car ces gènes pourraient influencer l'infection de cellules humaines par le Plasmodium. Nous avons ainsi observé la liaison génétique de certain SNP de ces gènes avec la parasitémie. Les anticorps IgG dirigés contre le parasite P. falciparum sont connus pour jouer un rôle important dans la réponse immunitaire. Certain gènes associés à la résistance au paludisme pourraient également être associés au niveau de production d'IgG. Nous avons montré que certains polymorphismes situés dans les gènes HBB, FcRIIA, et du TNF influencent le niveau de production de différentes sous classes IgG. Ces résultats nous permettent de mieux comprendre le contrôle génétique de la réponse humorale contre la malaria.J'ai aussi caractérisé l'inflammation non infectieuse dans un modèle murin afin de tester l'influence de l'état inflammatoire lors de l'infection par le plasmodium. L'étude transcriptomique de trois organes réalisée sur un modèle de souris injectée avec de l'acide oléique, nous a permis de mettre en évidence une réponse inflammatoire principalement au niveau des poumons et du cerveau. Ces résultats vont nous permettre de tester l'hypothèse qu'une réponse inflammatoire pré-existante favoriserait la survenue de forme sévère du paludisme chez des souris infectées par Plasmodium. / My thesis focused on the study of factors influencing the future of malaria infection. A genetic linkage study based on markers distributed throughout the genome has allowed us to highlight the linkage of 17p11 - p13 chromosome region with parasitaemia. We investigated the genetic variation of genes and HS3ST3A1 HS3ST3B1 human chromosome 17p12 because these genes may influence the infection of human cells by Plasmodium . We observed genetic linkage of some SNPs in these genes with parasitaemia.IgG antibodies against the parasite P. falciparum are known to play an important role in the immune response . Certain genes associated with resistance to malaria could also be associated with the level of IgG production . We have shown that some polymorphisms located in genes HBB , Fc  RIIA , and TNF influence the level of production of different IgG subclasses . These results allow us to better understand the genetic control of humoral response against malaria.I also characterized non- infectious inflammation in a mouse model to test the influence of the inflammatory condition during infection by Plasmodium . The transcriptomic study performed on three organs mouse model injected with oleic acid , has allowed us to demonstrate an inflammatory response mainly in the lungs and brain. These results will allow us to test the hypothesis that pre-existing inflammatory response favor the occurrence of severe malaria in mice infected with Plasmodium.

Les sarcomes pléomorphes sont des tumeurs mésenchymateuses rares caractérisées par de nombreux remaniements chromosomiques. Leur processus d’oncogenèse reste encore mal compris, aucune altération génétique motrice de l’initiation tumorale n’a pu être identifiée de façon récurrente et spécifique à ce jour. Les travaux que j’ai réalisés durant ma thèse avaient pour but de mieux comprendre la biologie de ces tumeurs, notamment les conséquences transcriptomiques de leur remodelage génomique.Nous avons caractérisé les transcrits de fusion exprimés dans ces tumeurs par séquençage haut-débit (RNA-seq). Ceci nous a amené à identifier l’expression de plusieurs transcrits chimériques impliquant le gène TRIO (5,1% des tumeurs). De plus, cette analyse ainsi que l’identification de variants exprimés nous ont permis d’identifier de fréquentes mutations de gènes suppresseurs de tumeurs tels qu’ATRX (16% des tumeurs) et plus généralement des membres du complexe SWI/SNF (47% des tumeurs). Les altérations de ce complexe majeur de remodelage de la chromatine sont associées à une plus grande instabilité génétique et à un phénotype plus agressif.Dans les sarcomes pléomorphes, l’instabilité génétique est liée à la progression tumorale via l’expression d’une signature transcriptomique pronostique. Cette signature, nommée CINSARC, est impliquée dans le contrôle de la mitose et de la ségrégation chromosomique. Nous avons voulu déterminer l’origine de cette expression via une étude intégrant la génomique et des mécanismes de régulation épigénétique, transcriptionnelle et post-transcriptionnelle. Si ces mécanismes ne semblent pas directement causals de l’expression de CINSARC, d’importantes modifications ont pu être mises en évidences. D’un point de vue clinique, nous avons démontré que l’expression de cette signature est un facteur pronostique universel de l’agressivité tumorale dans de nombreux types de cancers. De plus, CINSARC est un meilleur marqueur pronostique que le grade FNCLCC, actuel standard international d’évaluation du risque métastatique des sarcomes des tissus mous. Nous avons ainsi développé une méthode permettant une application clinique routinière de la signature CINSARC afin d’améliorer la prise en charge thérapeutique de ces tumeurs. / Pleomorphic sarcomas are rare mesenchymal tumors characterized by many chromosomal rearrangements. Their oncogenic process is still poorly understood, no recurrent and specific genetic alteration able to drive the tumor initiation has been identified yet. The work I did during my thesis had the objective to better understand the biology of these tumors, focusing on transcriptomic consequences of their genomic remodeling.We characterized fusion transcripts in these tumors by high-throughput sequencing (RNA-seq). This led us to identify the expression of several chimeric transcripts involving TRIO (5.1% of cases). Moreover, this analysis and the identification of expressed variants allowed us to identify frequent mutations of tumor suppressor genes such as ATRX (16% of cases) and more generally members of the SWI/SNF complex (47% of cases). Alterations of this major complex of chromatin remodeling are associated with higher genetic instability and more aggressive phenotype.In pleomorphic sarcomas, genetic instability is linked to tumor progression through the expression of a prognostic transcriptomic signature. This signature, termed CINSARC, is involved in mitosis control and chromosome segregation pathways. We wanted to determine the origin of such expression by integrating genomics and epigenetics, transcriptional and post-transcriptional regulation mechanisms. Though these mechanisms do not seem to directly regulate CINSARC expression, important changes have been highlighted. From a clinical standpoint, we demonstrated that the signature expression is a global prognostic factor of tumor aggressiveness in numerous cancer types. In addition, CINSARC is a better prognostic marker than the FNCLCC grading system, the current international standard to evaluate the metastatic risk in soft tissue sarcomas. We consequently developed a method allowing a daily clinical application of the CINSARC signature to improve the therapeutic management of these tumors.

L'épissage alternatif est un processus biologique qui génère la diversité du protéome malgré le nombre limité de gène. Ce mécanisme régule à la fois les gènes de manières qualitatives (isoformes exprimées) mais aussi quantitatives (niveau d'expression). Avec le développement des technologies de séquençage à haut débit, il est maintenant possible d'étudier à large échelle les aspects quantitatif et qualitatif du transcriptome avec une même expérience (RNA-seq). Durant ma thèse, j'ai développé une nouvelle méthode d'analyse de l'épissage alternatif dans les données RNA-seq. J'ai également participé à la mise en place du pipeline global d'analyse de données RNA-seq (expression et épissage) qui a été utilisé pour analyser un grand nombre de jeux de données. Dans un second temps, nous avons comparé notre outil d'analyse de l'épissage, FaRLine, qui est basé sur l'alignement sur un génome de référence, à KisSplice, une méthode basée sur l'assemblage. Nous avons montré que ces méthodes trouvaient un grand nombre d'événements en communs (70%), mais qu'il existait des différences non négligeables dues à la méthodologie. Nous avons analysé et classifié ces événements en 4 grandes catégories. Les variants faiblement exprimés et les exons chevauchant des éléments répétés sont mieux annotés par les méthodes basées sur l'alignement. Alors que les méthodes basées sur l'assemblage trouvent des nouveaux variants (exons ou sites d'épissage non annotés) et prédisent de l'épissage alternatif dans les famille de gènes paralogues. Notre travail souligne les points qui nécessitent encore l'amélioration des méthodes bioinformatiques. Enfin, j'ai participé au développement de méthodes permettant d'aider les biologistes à évaluer l'impact fonctionnel de modification d'épissage, que ce soit au niveau de la protéine produite (annotation des domaines protéiques au niveau des exons), ou à un niveau plus global en intégrant les modifications d'épissage dans les voies de signalisation / Alternative splicing is the biological process that explain the large diversity of the proteome compared to the limited number of genes. This process allow a qualitative regulation (expressed isoforms) and a quantitative regulation (expression level). The growth of high-trhoughtput sequencing methods enabled the analysis of these two aspects (quantitative and qualitative regulation) with the same experiment (RNA-Seq). During my PhD, I developped a new tool to analyse alternative splicing from RNA-Seq data. I also participated in the automatisation of the complet pipeline of RNA-Seq analysis (expression and splicing). This pipeline has been used to analyse various datasets. Then, we compared our mapping-first tool, FaRLine, with an assembly-first method, KisSplice. We found that the predictions of the two pipelines overlapped (70\% of exon skipping events were common), but with noticeable differences. The mapping-first approach allowed to find more lowly expressed splicing variants, and was better in predicting exons overlapping repeated elements. The assembly-first approach allowed to find more novel variants, including novel unannotated exons and splice sites. It also predicted AS in families of paralog genes. Our work point out where the bioinformatic improvment are still needed. Finally, I participated in the developpement of bioinformatics methods to help biologists to evualuate the fonctionnal impact of splicing alteration : at the level of the protein product by annotating fonctionnal domain at the exon level or at a more global level, by integrating splicing modifications in signaling pathways

The pathological mechanisms of variant Creutzfeldt-Jakob disease (vCJD) in the human brain remain poorly understood. Gene expression data may provide insight into the molecular mechanisms involved. This requires analysis of human postmortem brain tissue however; the variability in RNA preparations from human brain material is a concern. A method for the isolation of RNA from vCJD brains which minimized infectivity and reduced Proteinase K resistant prion protein levels to undetectable by biochemical assay was developed. RNA preparations were made from sample of the frontal parasagital cortex, sub-frontal cortex and cerebellum of 78 human autopsy cases; 21 vCJD, 26 other neurological disease (OND) and 31 nonneurological disease (NND). Suitable RNA metrics for these human brain RNA preparations were evaluated and the intra- and inter-case variability of RNA preparations was determined. There was marked intra- and inter-case variability in RNA integrity number (RIN), A260:280 absorbance ratio and RNA yield. In particular, RIN and A260:280 showed little variation intra-patient, although RNA yield was more variable. The effects of postmortem interval, tissue pH, age at death, gender, freeze-thaw cycles (including storage method and temperature) and agonal state were investigated; none of these parameters correlated with the marked variability observed. Parameters for matching vCJD and OND/NND cases were considered and RNA from three age and gender matched comparison groups, each containing one OND, one NND and one vCJD case, were used for gene expression analysis. Data was generated using Superarray GEArray® Focused DNA Microarray and analysed using the GEArray Expression Analysis Suite and Significance Analysis of Microarray software. A comparison between matched vCJD and NND control cases identified 26 up-regulated and 16 down-regulated genes, showing >1.5-fold change with a false discovery rate of 9%. The modulated genes were involved in cell signaling, cell death, cholesterol and lipid metabolism. Involvement of these pathways is consistent with findings in other transmissible spongiform encephalopathy studies.