The aim of this study was to characterize the ophthalmic and genetic features of Bardet Biedl (BBS) syndrome in a cohort of patients from a German specialized ophthalmic care center. Sixty-one patients, aged 5–56 years, underwent a detailed ophthalmic examination including visual acuity and color vision testing, electroretinography (ERG), visually evoked potential recording (VEP), fundus examination, and spectral domain optical coherence tomography (SD-OCT). Adaptive optics flood illumination ophthalmoscopy was performed in five patients. All patients had received diagnostic genetic testing and were selected upon the presence of apparent biallelic variants in known BBSassociated genes. All patients had retinal dystrophy with morphologic changes of the retina. Visual acuity decreased from ~0.2 (decimal) at age 5 to blindness 0 at 50 years. Visual field examination could be performed in only half of the patients and showed a concentric constriction with remaining islands of function in the periphery. ERG recordings were mostly extinguished whereas VEP recordings were reduced in about half of the patients. The cohort of patients showed 51 different likely biallelic mutations—of which 11 are novel—in 12 different BBS-associated genes. The most common associated genes were BBS10 (32.8%) and BBS1 (24.6%), and by far the most commonly observed variants were BBS10 c.271dup;p.C91Lfs*5 (21 alleles) and BBS1 c.1169T>G;p.M390R (18 alleles). The phenotype associated with the different BBS-associated genes and genotypes in our cohort is heterogeneous, with diverse features without genotype–phenotype correlation. The results confirm and expand our knowledge of this rare disease.

Répondant au besoin de mieux comprendre les relations qui lient un génotype aux phénotypes moléculaires et cliniques associés, nous avons développé une nouvelle infrastructure bioinformatique qui unit, dans un même système, la collecte, la gestion, la maintenance et le traitement de multiples données ou informations. La première contribution de cette thèse est SM2PH Central et sa capacité de générer des instances. SM2PH Central constitue notre centre de référence en ligne pour toutes les protéines humaines intégrant des niveaux d’informations qui vont des aspects génomiques, structuraux, fonctionnels ou évolutifs aux aspects de transcriptomique, interactomique, protéomique ou métabolomique. La deuxième contribution est MSV3d, une ressource d’annotation multi-niveau (propriétés physico-chimiques, fonction, évolution, structure) des mutations humaines connues. MSV3d fournit l’ensemble des connaissances exploitées par la troisième contribution de cette thèse à savoir KD4v, notre base d’extraction de connaissances pour prédire l’impact phénotypique d’une mutation. La base de connaissances de KD4v induite par la Programmation Logique Inductive contient des règles exploitables par un humain ou un ordinateur et des facteurs prédictifs caractérisant les mutations neutres ou délétères. Enfin, l’ultime contribution de cette thèse est liée au développement de GEPeTTO, un prototype de priorisation de gènes. Une application biologique a été réalisée. Nous avons étudié la cécité nocturne en utilisant SM2PH Central, en combinaison avec le service d’annotation de MSV3d et la méthode de prédiction KD4v pour analyser le gène GPR179 et ses deux mutations nouvellement identifiées. / Responding to the need to better understand the relationships linking the genotype to the molecular and clinical phenotype, we have developed a new bioinformatics infrastructure that unites, in a single system, the collection, the management, the maintenance and the processing of multiple data or information. The first contribution of this thesis is SM2PH Central and its ability to generate instances. SM2PH Central is our online reference center for all human proteins including many levels of information such as genomics, structural, functional and evolutionary aspects of transcriptomics, interactomics, proteomics or metabolomics. The second contribution is MSV3d, a multi-level annotation resource (physico-chemical properties, function, evolution, structure) of known human mutations. MSV3d provides the knowledge used by the third contribution of this thesis namely KD4v, our knowledgebase extraction to predict the phenotypic effect of a mutation. The KD4v knowledgebase computed by Inductive Logic Programming contains the rules describing the information that can be either exploited by a human or a computer, and the predictors characterizing neutral or deleterious mutations. The last contribution of this thesis is related to the development of GEPeTTO, a prototype of the prioritization of genes. Finally, these tools (SM2PH Central, MSV3d, KD4v) allowed us in the context of patients data analysis to confirm the implication of GPR179 as a new gene responsible for congenital stationary night blindness.

La déficience intellectuelle (DI) est souvent associée à d’autres signes cliniques morphologiques et psychiatriques mais cette comorbidité est peu caractérisée pour la DI associée à un gène donné. Ainsi les mutations du gène p21-activated kinase 3 (PAK3) sont responsable d’un large spectre clinique, allant de la DI légère à des DI sévères associées parfois à des malformations du cerveau. Nous avançons l’hypothèse que les différentes mutations d’un même gène peuvent affecter divers paramètres biochimiques et affecter de manière différentielle les voies de signalisation impliquées dans la plasticité synaptique et dans le développement du cerveau. Pour valider notre hypothèse, nous avons caractérisé une nouvelle mutation responsable d’une déficience intellectuelle sévère associée à une agénésie du corps calleux et une microcéphalie. Cette mutation supprime l’activité kinase, n’affecte pas la stabilité de la protéine et augmente l’interaction avec un GEF de la famille PIX. Ces derniers résultats identifient une nouvelle voie de signalisation impactée par certaines mutations de PAK3. L’expression de ce variant modifie la morphologie cellulaire et la dynamique des adhésions focales, ainsi que les propriétés migratoires des cellules, ce qui pourraient relier les défauts biochimiques aux défauts de certaines fonctions cellulaires. De manière intéressante, ces caractéristiques sont aussi retrouvées pour un autre variant responsable d’une clinique également très sévère. Nous avons également caractérisé d’autres mutations associées à des phénotypes moins sévères. La synthèse de nos résultats nous permet ici de proposer un modèle explicatif de la relation génotype-phénotype pour les mutations de déficience intellectuelle liées au gène PAK3, intégrant des défauts neurodéveloppementaux et de plasticité synaptique. / Intellectual Disability (ID) is often associated with other morphological and psychiatric clinical signs, but this comorbidity is poorly characterized for ID associated with a given gene. Thus mutations of the p21-activated kinase 3 (PAK3) gene are responsible for a broad clinical spectrum, ranging from mild ID to severe ID, sometimes associated with brain malformations. We hypothesize that different mutations of the same gene may affect various biochemical parameters and differentially affect the signaling pathways involved in synaptic plasticity and brain development. To validate our hypothesis, we characterized a new mutation responsible for a severe intellectual disability associated with agenesis of the corpus callosum and microcephaly. This mutation suppresses kinase activity, does not affect protein stability and increases the interaction with a GEF of the PIX family. These latest results identify a new signaling pathway impacted by certain PAK3 mutations. The expression of this variant modifies the cellular morphology and the dynamics of the focal adhesions, as well as cell migratory properties, which could link the biochemical defects to those of certain cell functions. Interestingly, these features are also found for another variant responsible for a very similar severe clinical spectrum. We have also characterized other mutations associated with less severe phenotypes. The synthesis of our results allows us to propose an explanatory model of the genotype-phenotype relationship integrating neurodevelopmental and synaptic plasticity defects for intellectual disability and other clinical traits associated to the PAK3 gene mutations.

Most genome-wide association studies search for genetic variants associated to a single trait of interest, despite the main interest usually being the understanding of a complex genotype-phenotype network. Furthermore, many studies collect data on multiple phenotypes, each measuring a different aspect of the biological system under consideration, therefore it can often make sense to jointly analyze the phenotypes. However this is rarely the case and there is a lack of well developed methods for multiple phenotype analysis. Here we propose novel approaches for genome-wide association analysis, which scan the genome one SNP at a time for association with multivariate traits. The first half of this thesis focuses on an analytic model averaging approach which bi-partitions traits into associated and unassociated, fits all such models and measures evidence of association using a Bayes factor. The discrete nature of the model allows very fine control of prior beliefs about which sets of traits are more likely to be jointly associated. Using simulated data we show that this method can have much greater power than simpler approaches that do not explicitly model residual correlation between traits. On real data of six hematological parameters in 3 population cohorts (KORA, UKNBS and TwinsUK) from the HaemGen consortium, this model allows us to uncover an association at the RCL locus that was not identified in the original analysis but has been validated in a much larger study. In the second half of the thesis we propose and explore the properties of models that use priors encouraging sparse solutions, in the sense that genetic effects of phenotypes are shrunk towards zero when there is little evidence of association. To do this we explore and use spike and slab (SAS) priors. All methods combine both hypothesis testing, via calculation of a Bayes factor, and model selection, which occurs implicitly via the sparsity priors. We have successfully implemented a Variational Bayesian approach to fit this model, which provides a tractable approximation to the posterior distribution, and allows us to approximate the very high-dimensional integral required for the Bayes factor calculation. This approach has a number of desirable properties. It can handle missing phenotype data, which is a real feature of most studies. It allows for both correlation due to relatedness between subjects or population structure and residual phenotype correlation. It can be viewed as a sparse Bayesian multivariate generalization of the mixed model approaches that have become popular recently in the GWAS literature. In addition, the method is computationally fast and can be applied to millions of SNPs for a large number of phenotypes. Furthermore we apply our method to 15 glycans from 3 isolated population cohorts (ORCADES, KORCULA and VIS), where we uncover association at a known locus, not identified in the original study but discovered later in a larger one. We conclude by discussing future directions.

A hiperplasia adrenal congênita (HAC) é um grupo de doenças hereditárias causadas por uma deficiência em uma das enzimas necessárias para a síntese de cortisol no cortex adrenal. Mais de 95% de todos os casos de HAC são devidos a 21-Hidroxilase (21-OHD). Existem 3 formas principais, duas com manifestações clínicas no período neonatal, a forma mais grave perdedora de sal (HAC-PS) e a forma virilizante simples (HAC-VS). Além da perda salina, o excesso de andrógenos leva à virilização de recém nascidas femininas. As formas neonatais são chamadas de clássicas, atividade enzimática da 21-OH bastante reduzida, de < 2% na HAC-PS e de 2-10% na HAC-VS. A forma parcial de início tardio é chamada de HAC não clássica (HAC-NC) e a principal manifestação na infância é a adrenarca precoce. Nesta forma a atividade da 21-OH é de 20-60%. Os programas de triagem para HAC visam, principalmente, ao diagnóstico precoce da forma clássica perdedora de sal, mais grave e potencialmente letal. No Brasil, a triagem pública é realizada no Estado de Goiás desde 1997 e em Santa Catarina desde 2001. No Rio Grande do Sul (RS) foi implantada em maio de 2014 na fase IV do Programa Nacional de Triagem Neonatal. A inclusão da HAC trouxe consigo muitos desafios e a exigência de um fluxo de triagem e diagnóstico bem estruturados. O diagnóstico precoce é crucial para prevenir o óbito de lactentes por insuficiência adrenal. Dosa-se, em papelfiltro, a 17OH progesterona (17-OHP). Elevações podem ocorrer em recém-nascidos sem HAC (falso-positivos), devido a situações de estresse perinatal e por prematuridade. Após avaliação dos dados do primeiro ano de triagem para HAC neste estudo, a mediana da idade da coleta nos casos diagnosticados foi de 8 dias (4.25-15.75). Dos 8 casos diagnosticados de maio de 2014 a abril de 2015, 6 casos com forma perdedora de sal (incluindo 1 caso de óbito por coleta tardia do TP aos 38 dias de vida). A incidência encontrada em nosso meio no primeiro ano foi de 1:13.551 casos. Com a estratégia do uso de pontos de corte estratificados pelo peso de nascimento18, o índice total de resultados positivos em nosso meio foi de 0,5% da amostra avaliada (“n” total de 514 bebês), sendo mais frequente em recém nascidos com menos de 2000g de peso ao nascer. Além da confirmação clínica e laboratorial, o genótipo é importante, além de confirmar, para estabelecer gravidade da doença e também para ratificar o diagnostico dos falsos positivos na ausência de uma mutação do gene CYP21A2. Um dos casos confirmados de HAC-PS foi associado a múltiplas malformações e craniossinostose severa, suscitando a hipótese de um a associação com defeito de FGFR2. A correlação genótipo- fenótipo na avaliação dos casos em dois anos da triagem alcançou um alto nível de concordância de 87%. Diagnosticada, portanto, de forma assertiva a HAC forma clássica, instituise a terapia glicocorticóide para as formas virilizante simples e acrescenta-se mineralocorticóide para as formas perdedoras de sal. A triagem neonatal é um importante programa de saúde populacional e visa ao diagnóstico precoce de uma patologia com potencial risco à vida pela perda de sal, além de permitir adequada atribuição de sexo nas meninas com virilização genital e à saúde da criança. Estabelecer os fluxos adequados de triagem e manejo, além de ampliar o conhecimento sobre a HAC, com o reconhecimento dos desfechos e tratamentos adequados é essencial para minimizar as possíveis complicações nesta população de maior vulnerabilidade. / Congenital adrenal hyperplasia (CAH) is a group of inherited diseases caused by a deficiency in one of the enzymes required for the cortisol synthesis by the adrenal cortex. More than 95% of all CAH cases are due to 21-hydroxylase (21-OHD). There are 3 forms, two with neonatal clinical manifestation: salt-wasting CAH (SW-CAH\) and the simple virilizing form (HAC-VS). In addition to salt loss, androgens excess lead to the virilization of female newborn. Neonatal forms are defined as classical CAH. The 21-OHD enzymatic activity in SWCAH is less than <2% and in the SV-CAH 2-10%. A late-onset form, with partial enzymatic defect (20-60%) is called non-classical HAC (NC-CAH) and the main manifestation in childhood is early adrenarche. In Brazil, public health screening has been conducted in the State of Goiás since 1997 and in te Sate of Santa Catarina since 2001. In Rio Grande do Sul (RS) it was implemented in May 2014, in phase IV of the National Neonatal Screening Program. The inclusion of CAH in the local screening program brought many challenges and the need of a well structured screening and diagnosis flowchart. Early diagnosis is crucial to prevent infant death due to adrenal insufficiency. Around de world, the screening programs for CAH main purpose is the early diagnosis of the more severe classical forms, especially SW-CAH. The cortisol precursor 17OH progesterone (17-OHP) is the main disease marker and is measured on filter paper. Elevations may occur in infants without CAH (false positive) due to perinatal stress and prematurity. Of newborns screened in the first year, median age of collection in diagnosed cases was 8 days (4.25-15.75) and 8 patients were diagnosed with CAH (four males, four females). The incidence of CAH in the state was 1:13,551. Six cases were identified as classic salt-was-ting CAH and two were cases of virilizing CAH. The overall rate of positive results was 0.5% (n = 514 infants). The number of false positive results was higher among newborns with birth weight < 2,000 g. In addition to clinical and laboratory confirmation, the genotype is important to confirme 21-OH deficiency, to establish disease severity and also in the absence of a mutation of the CYP21A2 gene to more precise exclude the diagnosis of suspected false positives. One of the confirmed cases of SW-CAH was associated with multiple malformations and severe craniosynostosis, raising the hypothesis of an association with FGFR2 mutation. A high genotype- phenotype correlation of 87% was found in the cases after two years of screening. Once the classic CAH is diagnosed, glucocorticoid therapy is instituted and mineralocorticoid is added for SW-CAH. CAH neonatal screening is an important population health program and aims to the early diagnosis of a pathology with a potential risk due to salt loss crisis. The early detection of cases also allows to adequate sex assignment in girls with genital virilization. Establishing adequate screening flows, proper diagnosis and management, in addition to increase knowledge about the disease, with the appropriate recognition of outcomes and treatments is essential to minimize complications in this population of greater vulnerability.

A l’ère des omiques, l’étude des relations génotype-phénotype repose sur l’intégration de données diverses décrivant des aspects complémentaires des systèmes biologiques. La génomique comparative offre un angle d’approche original, celui de l’évolution, qui permet d’exploiter la grande diversité phénotypique du Vivant. Dans ce contexte, mes travaux de thèse ont porté sur la conception de marqueurs évolutifs décrivant les gènes selon leur histoire évolutive. Dans un premier temps, j’ai construit une ressource d’orthologie complète, OrthoInspector 3.0 pour extraire une information évolutive synthétique des données génomiques. J’ai ensuite développé des outils d’exploration de ces marqueurs en relation avec les données fonctionnelles et/ou phénotypiques. Ces méthodes ont été intégrées à la ressource OrthoInspector ainsi qu’au réseau social MyGeneFriends et appliquées à l’étude des ciliopathies, conduisant à l’identification de 87 nouveaux gènes ciliaires. / In the omics era, the study of genotype-phenotype relations requires the integration of a wide variety of data to describe diverse aspects of biological systems. Comparative genomics provides an original perspective, that of evolution, allowing the exploitation of the wide phenotypic diversity of living species. My thesis focused on the design of evolutionary markers to describe genes according to their evolutionary history. First, I built an exhaustive orthology resource, called OrthoInspector 3.0, to extract synthetic evolutionary information from genomic data. I then developed methods to explore the markers in relation to functional or phenotypic data. These methods have been incorporated in the OrthoInspector resource, as well as in the MyGeneFriends social network and applied to the study of ciliopathies, leading to the identification of 87 new ciliary genes.