Premature ovarian Insufficiency (POI) is the cessation of the ovarian function before the age of 40, defined by high serum gonadotrophins, low estradiol and amenorrhea for at least 4 months. The etiology may be iatrogenic after a surgery, chemotherapy or radiotherapy treatment, environmental, autoimmune or genetic. However, in most of the cases the cause remains unknown. Clinical and family studies suggest a strong heritability of age at menopause and POI, but the number of genetic causes and genes identified to be involved in human POI remains very small. In POI patients, the two crucial functions of the ovary, hormonal secretion and reproduction, are absent. In the last decades, however, new advances in assisted reproduction techniques have allowed the possibility of carrying pregnancies to POI women, thanks to oocyte donation. The aim of this study was to identify new genetic factors implicated in the development of POI women and to analyse the reproductive possibilities and outcome of women with a genetic cause of POI. For the first part of the study, the DNA of a cohort of POI women recruited in the Fertility Clinic of the Hôpital Erasme of the Université Libre de Bruxelles was used to sequence five candidate genes (FSHR, GDF9, BMP15, AMH and AMHR2) known to be implicated in the ovarian folliculogenesis. The most important findings were two very rare variants and one unknown variant in the AMH gene. The functional study performed with these variants suggested a diminished function of the mutant protein. Furthermore, one of the variants was found in the mother of one of the patients, who was also diagnosed of POI at 32 years old. These arguments strongly suggest that a defective AMH could play a role in the development of POI. This is supported by previous studies with knock out mice models, which show an earlier depletion of the ovarian follicle pool due to a faster recruitment of the primordial follicles that constitute the ovarian reserve. The sequencing of the BMP15 gene led to the identification of two new variants not identified among controls but not predicted to be deleterious. Interestingly, one variant previously reported in POI women and predicted to be deleterious for the protein function, was found in a Sub-Saharan African POI patient as well as in our control cohort. This variant was already studied functionally and shown to have a reduced biological activity. However, we identified this variant in 6% of the Sub-Saharan African control population, which suggests that this is a more prevalent variant in the African genotype and raises up the importance of the ethnicity when studying genetic variants.The sequencing of the other genes (FSHR, GDF9 and AMHR2) did not lead to any association with POI.In the second part of the study, 24 women with Turner syndrome and POI were analysed in terms of reproductive, obstetrical and perinatal outcome after oocyte donation. This specific group of patients was chosen because of their specific systemic anomalies that could interfere with pregnancy outcome and because very few reports have been published on this subject. In the 23 patients finally transferred, the pregnancy rate was similar to that obtained after oocyte donation in other cohorts. There was a miscarriage rate of 23% and a rate of complications of pregnancy as high as 50%, mainly caused by pregnancy-induced hypertensive diseases. Four women at risk of genetic POI were included in the fertility preservation program in order to vitrify their oocytes. Three of them have already vitrified successfully their oocytes but none of them has yet used them.Oocyte vitrification represents a new hope for those women with genetic risk of POI to be able to carry a pregnancy with their own oocytes.In conclusion, three variants of the AMH gene could be implicated in the development of POI as demonstrated by the reduced in vitro bioactivity of the variants and the familial segregation of the cases. Since then, it sounds plausible to propose AMH sequencing in the case of familial POI and secondary amenorrhea.In the BMP15 gene, 2 new variants were predicted to be tolerated. A potentially deleterious variant of the BMP15 gene (L148P) previously associated to POI, was also found in 6% of the Sub-Saharan African controls which suggests that it is a common variant in the African ethnic. No clear association was found between the other tested candidate genes and our POI cohort.Regarding Turner’s Syndrome pregnancies, we can conclude that they are high-risk pregnancies that need of a multidisciplinary follow-up before and during pregnancy.Oocyte cryopreservation represents a new tool to be offered to women at risk of genetic POI to preserve their fertility, but not without previous genetic counselling. / Doctorat en Sciences médicales (Médecine) / info:eu-repo/semantics/nonPublished

The emergence of statistical and predictive methods able to analyse genomic data has revolutionised the field of medical genetics, allowing the identification of disease-causing gene variants (i.e. mutations) for several human genetic diseases. Although these approaches have greatly improved our understanding of Mendelian «one gene – one phenotype» genetic models, studying diseases related to more intricate models that involve causative variants in several genes (i.e. oligogenic diseases) still remains a challenge, either due to the lack of sufficient methodologies and disease-specific cohorts to study or the phenotypic complexity associated with such diseases. This situation makes it difficult to not only understand the genetic mechanisms of the disease, but to also offer proper counseling and support to the patient. Until recently, no specialized predictive methods existed to directly predict causative variant combinations for oligogenic diseases. However, with the advent of data on variant combinations in gene pairs (i.e. bilocus variant combinations) leading to disease, collected at the Digenic Diseases Database (DIDA), we hypothesized that the transition from single to variant combination pathogenicity predictors is now possible.To investigate this hypothesis, we organised our research on two main routes. At first, we developed an interpretable variant combination pathogenicity predictor, called VarCoPP, for gene pairs. For this goal, we trained multiple Random Forest algorithms on pathogenic bilocus variant combinations from DIDA against neutral data from the 1000 Genomes Project and investigated the contribution of the incorporated variant, gene and gene pair features to the prediction outcome. In the second part, we explored the usefulness of different gene pair burden scores based on this novel predictive method, in discovering oligogenic signatures in neurodevelopmental diseases, which involve a spectrum of monogenic to polygenic cases. We performed a preliminary analysis on the Deciphering Developmental Diseases (DDD) project containing exome data of 4195 families and assessed the capability of our scores in supporting already diagnosed monogenic cases, discovering significant pairs compared to control cases and linking patients in communities based on the genetic burden they share, using the Leiden community detection algorithm.The performance of VarCoPP shows that it is possible to predict disease-causing bilocus variant combinations with good accuracy both during cross-validation and when testing on new cases. We also show its relevance for disease-related gene panels, and enhanced its clinical applicability by defining confidence zones that guarantee with 95\% or 99\% probability that a prediction is indeed a true positive, guiding clinical researchers towards the most relevant results. This method and additional biological annotations are incorporated in an online platform called ORVAL that allows the prediction and exploration of candidate disease-causing oligogenic variant combinations with predicted gene networks, based on patient variant data. Our preliminary analysis on the DDD cohort shows that - although all bi-locus burden scores show advantages, disadvantages and certain types of biases - taking the maximum pathogenicity score present inside a gene pair seems to provide, at the moment, the most unbiased results. We also show that our predictive methods enable us to detect patient communities inside DDD, based exclusively on the shared pathogenic bi-locus burden between patients, with more than half of these communities containing enriched phenotypic and molecular pathway information. Our predictive method is also able to bring to the surface genes not officially known to be involved in disease, but nevertheless, with a biological relevance, as well as a few examples of potential oligogenicity inside the network, paving the way for further exploration of oligogenic signatures for neurodevelopmental diseases. / Doctorat en Sciences / info:eu-repo/semantics/nonPublished

This thesis aimed to establish how different HLA genotypes correlate to MS severity and disease progression and whether they could be used as additional disease biomarkers and to a large extent the work has succeeded in this task. Association of MS with the alleles HLA-DRB1*15 and HLA-DQB1*06 and haplotype DRB1*15-DQB1*06 was identified, and under representation of other alleles, such as the HLA-DRB1*07 and HLA-A*02 alleles, showed a potentially protective role against the disease. HLA-A*02 was shown to be a marker of a better prognosis and, in contrast, HLA-B*07, B*08 and B*44 seem to be associated to with a worse prognosis. / Doctorat en Sciences médicales (Médecine) / info:eu-repo/semantics/nonPublished

La majeure partie de l'héritabilité des maladies inflammatoires des intestins (MICI) comprenant essentiellement la maladie de Crohn (MC) et la rectocolite ulcéro-hémorragique (RCUH) reste manquante. A côté des formes classiques de MICI polygéniques, un nombre croissant de maladies rares monogéniques avec un syndrome d’immunodéficience primaire innée ou acquise (ou Primary Immune Deficiency (PID) disorders) et un tableau clinique de maladie inflammatoire gastro-intestinale semblable aux MICI ont été identifiées au cours de ces dernières années. Les gènes mutés au sein de ces PID pourraient contenir des variants rares associés aux MICI dont l'identification aurait échappé aux études d'association du génome (GWAS) et qui pourraient être détectés avec le séquençage à haut débit. Nous avons sélectionné 23 gènes candidats associés à dix maladies monogéniques avec manifestations gastro-intestinales inflammatoires ressemblant aux MICI et avons évalué la contribution éventuelle de ces gènes à la prédisposition héréditaire à développer une MC via d'une part une étude d'association sur base des données de génotypage de 17000 patients atteints de MC issus des GWAS de l'IIBDGC et d'autre part, via une étude de séquençage à haut débit au sein d'une cohorte de 2390 patients atteints de MC. Nous avons identifié 4 nouveaux variants du gène XIAP qui sont associés chez les patients porteurs de la mutation, à un défaut d'activation de la voie de NOD2 après stimulation par le muramyl dipeptide (MDP). En complément à l'étude des formes monogéniques de MICI, les formes non européennes peuvent également constituer une source de données à exploiter pour pallier à l'héritabilité manquante. En effet, la plupart des GWAS ont été effectuées au sein de populations d'origine européenne. Or en combinant plusieurs groupes ethniques, il est possible d'identifier de nouveaux loci de susceptibilité tel qu'on a pu l'observer dans la première étude trans-éthnique réalisée au sein de patients atteints de MICI. Très peu d'études génétiques ont été réalisées au sein de la population d'origine marocaine qui est la population non européenne la plus prépondérante en Belgique. Nous avons entrepris une étude d'association sur des puces de génotypage de type Immunochip au sein d'une population de 306 patients d'origine marocaine atteints de MICI, se focalisant sur les 163 variants communs de susceptibilité connus dans les MICI lorsque l'étude a été débutée. L'étude a identifié une association significative aux MICI pour dix loci notamment au sein des gènes tels que IL23R, JAK2, CARD9 ou CCR6. Ces associations sont nouvelles et n'ont pas été étudiées ou confirmées dans les précédentes études faites au sein de patients d'origine marocaine atteints de MICI. / Doctorat en Sciences médicales (Médecine) / info:eu-repo/semantics/nonPublished

Cette étude pilote vise à développer une méthode d'analyse intégrative qui permet d'augmenter le taux de réussite du diagnostic clinique des mutations génétiques rares. De plus, l'identification de nouveaux gènes associés à l'ataxie épisodique (EA) et l'évaluation de nouveaux algorithmes de prédiction, pour un examen de variants plus robuste, découleront de l'enquête. Caractérisé par une perte sporadique de la coordination des mouvements volontaires, l'EA se manifeste généralement tardivement, avec une hétérogénéité clinique et génétique élevée, compliquant largement l’obtention d’un diagnostic précis. Alors que quatre gènes ont été liés aux huit sous-types d'EA, de nombreux patients demeurent sans diagnostic moléculaire dû aux limites des méthodes de séquençage d’ADN. Ces lacunes accentuent l’intérêt d’implanter le séquençage de l’ARN en milieu clinique, afin d’obtenir l’information fonctionnelle offerte par l’approche. Des patients atteints d’EA, sans diagnostic moléculaire malgré un examen approfondi, ont été recrutés à Montréal. Le séquençage du génome entier (WGS) et de l'ARN a été effectué sur des échantillons de sang pour identifier les variants nucléotidiques, l'expression différentielle, les événements d'épissage ainsi que les expansions de microsatellites. Plusieurs algorithmes de prédiction de la pathogénicité récents ont été choisis pour être testés parallèlement aux algorithmes standard. Des données WGS provenant d’un trio familial atteint de pathologies neurologiques ont également été soumises au pipeline génomique développé pour la cohorte EA. Des variants candidats ont été identifiés pour chaque patient en fonction des scores de pathogénicité, de la rareté des événements génétiques et des informations fonctionnelles et cliniques connues pour un gène altéré donné. Parmi les découvertes figurent des mutations non-sens, des faux-sens, de l'épissage alternatif ainsi que des expansions nucléotidiques dans des gènes associés aux ataxies spinocérébelleuses ou aux paraplégies spastiques. En plus d'être présents dans les ensembles de données de séquençage disponibles pour chaque patient, les événements génomiques ont été vérifiés par séquençage Sanger de l'ADN et de l'ARN lorsque possible. Les effets fonctionnels potentiels, prédits principalement à partir du RNA-seq et suggérant une expression anormale de l'ARNm, ont également été évalués par amplification PCR et qPCR traditionnelle. À ce jour, quatre des dix patients ont reçu ou sont en voie de recevoir un diagnostic clinique, et quatre autres présentent d’excellents candidats moléculaires pour expliquer une pathologie ataxique. Ce projet devrait permettre un diagnostic mieux défini, conduisant à une meilleure qualité de vie, une meilleure évaluation du pronostic et une meilleure prise en charge des patients. L’identification de modulateurs génétiques chez certains d’entre eux devrait également permettre une meilleure caractérisation clinique des conditions rapportées, bénéficiant les évaluations symptomatiques futures. De plus, la méta-analyse des données RNA-seq offre le potentiel de découvrir des régulateurs de pathogenèse communs à l’EA. Il favorisera également l'approche intégrative pour un plus large éventail de troubles et pourrait éventuellement conduire à de nouvelles stratégies thérapeutiques. / This pilot study aims to develop an integrative analysis method that allows for an increased diagnosis success rate of rare genetic mutations. Moreover, identification of novel genes associated with Episodic Ataxia (EA) and evaluation of new AI-generated prediction algorithms, for a more robust variant examination, will ensue from the investigation. Characterized by sporadic loss of voluntary movement coordination, EA typically manifest with a late onset as well as high-clinical and genetic heterogeneity, setting additional hurdles to diagnosis. While four genes have been linked to the eight subtypes of EA, many patients are left without molecular diagnosis due to the limitations of individual DNA-sequencing methods, which can be mitigated by the functional overview that RNA sequencing (RNA-seq) offers. EA patients, lacking molecular diagnosis despite in-depth examination, were recruited in Montreal. Whole-Genome sequencing (WGS) and RNA-seq were performed on blood samples to identify single nucleotide variants, differential expression, splicing events, structural variants and repeat expansions. Multiple recent pathogenicity prediction algorithms were chosen for testing concurrently to standard ones, in order to evaluate their performance and potential for clinical pipelines integration. WGS data of a family trio from France, in which the father and the daughter present neurologic pathologies, were also processed through the genomic pipeline that was developed for the EA cohort in order to identify the cause of their disorder. Candidate variants were identified for each patient according to pathogenicity scores, rarity of genetic events, and known functional as well as clinical information for a given altered gene. Among the findings are truncations, missenses, alternative splicing, and repeat expansions in genes already associated to either spinocerebellar ataxia or spastic paraplegia. In addition to being present in both datasets when available, validation of these interesting genomic events has been performed through Sanger Sequencing of both DNA and RNA when feasible. For strong candidates where the available functional information from RNA-seq suggests abnormal mRNA expression, validation includes PCR amplification as well as a traditional qPCR to support effects on transcripts. To this day, four out of ten patients have received or are on the verge of receiving a diagnosis, and four others are carrying excellent molecular candidates requiring further validation to explain their ataxic pathologies. This project should provide more defined diagnosis, leading to better quality of life, better evaluation of prognosis and better management of care for patients. Identification of genetic modifier in some of them should also allow for a better clinical characterization of the reported conditions, benefiting future patient examinations. A meta-analysis of our patients’ transcriptomic profiles could also uncover commonly affected pathways in EA development. It will also promote the integrative approach for a larger spectrum of disorders and might eventually lead to new therapeutic strategies.