Lolium perenne is one of the most important forage crops globally and in New Zealand. Lotus corniculatus is a dicotyledonous forage that contains valuable traits such as high levels of condensed tannins, increased digestibility, and high nitrogen fixing abilities. However, conventional breeding between these two forage crops is impossible due to their markedly different taxonomic origin. Protoplast fusion (somatic hybridisation) provides an opportunity for gene introgression between these two species. This thesis describes the somatic hybridisation, the regeneration and the molecular analysis of the putative somatic hybrid plants obtained between L. perenne and L. corniculatus. Callus and cell suspensions of different cultivars of L. perenne were established from immature embryos and plants were regenerated from the callus. Of the 10 cultivars screened, cultivars Bronsyn and Canon had the highest percentage of callus induction at 36% each on 5 mg/L 2,4-D. Removal of the palea and lemma which form the seed coat was found to increase callus induction ability of the embryos. Plant regeneration from the callus was achieved when the callus was plated on LS medium supplemented with plant growth regulators at different concentrations. Variable responses to shoot regeneration was observed between the different cultivars with the cv Kingston having the lowest frequency of shoot formation (12%). Different factors affecting the protoplast isolation of L. perenne were investigated. The highest protoplast yield of 10×10⁶ g⁻¹FW was obtained when cell suspensions were used as the tissue source, with enzyme combination 'A' (Cellulase Onozuka RS 2%, Macerozyme R-10 1%, Driselase 0.5%, Pectolyase 0.2%), for 6 h incubation period in 0.6 M mannitol. Development of microcolonies was only achieved when protoplasts were plated on nitrocellulose membrane with a L. perenne feeder layer on PEL medium. All the shoots regenerated from the protoplast-derived calli were albino shoots. The highest protoplast yield (7×10⁶ g⁻¹FW) of L. corniculatus was achieved from cotyledons also with enzyme combination 'A' (Cellulase Onozuka RS 2%, Macerozyme R-10 1%, Driselase 0.5%, Pectolyase 0.2%), for 6 h incubation period in 0.6 M mannitol. The highest plating efficiency for L. corniculatus of 1.57 % was achieved when protoplasts were plated on nitrocellulose membrane with a L. perenne feeder layer on PEL medium. The highest frequency of shoot regeneration (46%) was achieved when calli were plated on LS medium with NAA (0.1 mg/L) and BA (0.1 mg/L). Protoplast fusion between L. perenne and L. corniculatus was performed using the asymmetric somatic hybridisation technique using PEG as the fusogen. L. perenne protoplasts were treated with 0.1 mM IOA for 15 min and L. corniculatus protoplasts were treated with UV at 0.15 J/cm² for 10 min. Various parameters affecting the fusion percentage were investigated. Successful fusions were obtained when the fusions were conducted on a plastic surface with 35% PEG (3350 MW) for 25 min duration, followed by 100 mM calcium chloride treatment for 25 min. A total of 14 putative fusion colonies were recovered. Shoots were regenerated from 8 fusion colonies. Unexpectedly, the regenerated putative hybrid plants resembled L. corniculatus plants. The flow cytometric profile of the putative somatic hybrids resembled that of L. corniculatus. Molecular analysis using SD-AFLP, SCARs and Lolium specific chloroplast microsatellite markers suggest that the putative somatic hybrids could be L. corniculatus escapes from the asymmetric protoplast fusion process. This thesis details a novel Whole Genome Amplification technique for plants using Strand Displacement Amplification technique.

Les Wolbachia sont des alpha-proteobactéries présentes chez de nombreux arthropodes et nématodes filaires. Ces bactéries héritées maternellement induisent chez leurs hôtes des phénotypes allant du parasitisme au mutualisme, avec le long de ce continuum des phénotypes tels que la féminisation (F), l'incompatibilité cytoplasmique (IC) ou la mort des mâles. Wolbachia est ainsi un modèle particulièrement intéressant pour étudier les différents types de relations symbiotiques.Chez Brugia malayi, comme pour les autres nématodes filaires, Wolbachia vit en symbiose obligatoire avec son hôte. L'élimination de la bactérie par des traitements antibiotiques entraîne une perte de fertilité voire la mort du nématode. Chez l'isopode terrestre Armadillidium vulgare, Wolbachia induit la féminisation des mâles génétiques en femelles fonctionnelles entraînant des biais de sex-ratio vers les femelles dans la descendance.Pour comprendre les mécanismes impliqués dans ces deux symbioses, nous avons mis au point une nouvelle méthode de capture pour isoler l'ADN de Wolbachia et séquencer 8 souches de Wolbachia d'isopodes (F et IC). Une étude de génomique comparative a permis d'établir un premier pan-génome des bactéries du genre Wolbachia et d'identifier 2, 5 et 3 gènes présents seulement chez les souches mutualistes, féminisantes ou induisant la mort des mâles. L'expression des gènes potentiellement impliqués dans la féminisation ou le mutualisme a été étudiée au cours du développement de l'hôte. L'étude de l'interactome protéique bactérie-hôte a ensuite été initiée en utilisant comme appât des protéines bactériennes à domaines eucaryotes en vue d'identifier les cibles de Wolbachia chez l'hôte. / Bacteria of the genus Wolbachia are gram-negative alpha-proteobacteria present in many arthropods and filarial nematodes. These obligate intracellular bacteria are maternally inherited and induce a large number of phenotypes across the symbiosis continuum from mutualism to parasitism, including feminization (F), cytoplasmic incompatibility (CI) or male killing. Studying Wolbachia symbioses is therefore of particular interest in the investigation of symbiotic relationships.In Brugia malayi and other filarial nematodes, they are obligate leading to a loss of worm fertility, and eventual death upon their depletion with antibiotic. In arthropods, they rather are parasitic. In the isopod crustacean Armadillidium vulgare they cause feminization when present: genetic males develop as functional female leading to female biased sex-ratio progenies.In order to understand the molecular mechanisms of these two symbioses, we set up a new capture procedure to catch Wolbachia DNA and performed whole-genome sequencing on 8 Wolbachia strains, symbionts of isopods (F & CI). Comparative genomics led to the establishment of the Wolbachia pan-genome as well as the identification of phenotype related gene patterns. We identified 2, 5 and 3 genes that are only found in mutualist, feminizing and male killing strains, respectively. Expression of genes potentially involved in feminization and mutualism were also analyzed throughout host post-embryonic development. Host-symbiont interactome approach was then initiated by protein-protein interaction studies using bacterial proteins with eukaryote like motifs as bait in order to identify Wolbachia host targets involved in symbiosis.

No description available.

Les invasions biologiques dans les agroécosystèmes engendrent de lourdes pertes économiques. Parmi les nombreuses espèces en cause, on retrouve les nématodes phytoparasites, vers microscopiques s’attaquant principalement aux racines. Présent dans tous les principaux pays producteurs de soya, le nématode à kyste du soya (NKS), Heterodera glycines, serait à lui seul responsable annuellement de plusieurs milliards de dollars de pertes. La rotation avec des cultivars résistants est le moyen le plus efficace de contrôler les populations de NKS, mais la surutilisation des mêmes lignées a conduit à la sélection d’individus virulents et mené à leur inefficacité. À ce jour, les mécanismes ainsi que les gènes de virulence associés au contournement de la résistance continuent de mystifier les scientifiques. Dans cette thèse, les effets des changements climatiques sur la reproduction et l’établissement du NKS ainsi que sur la phénologie de son hôte, le soya, ont été étudiés. Le premier modèle bioclimatique simulant le cycle de vie du NKS et du soya a été développé. Il a démontré que le nématode peut déjà se reproduire dans toutes les régions du Québec et que la hausse attendue des températures dans le futur proche (2041-2070) permettrait au NKS de pratiquement doubler le nombre de générations produites par saison de croissance dans toutes les régions. De plus, la production de soya issu du groupe de maturité I pourrait s’étendre à toutes les régions du Québec d’ici 2070. Une étude sur la distribution de la variabilité génétique entre 64 populations américaines et ontariennes et les gènes associés à diverses composantes bioclimatiques et leur rôle dans l’adaptation a également été réalisée. Celle-ci a révélé que la diversité génétique était très élevée entre les populations et qu’un flux de gène continu aurait facilité l’adaptation du NKS à diverses conditions bioclimatiques et son établissement dans toutes les régions nord-américaines où l’on produit du soya. Finalement, cette thèse présente l’analyse des génotypes du NKS et des gènes différentiellement exprimés sur des plants de soya résistant (Peking et PI88788) et sensible (Essex). En plus d’identifier plusieurs protéines liées à la virulence, cette étude a permis de mettre en évidence une région génomique sous forte pression évolutive. Cet îlot génique contient plusieurs répétitions en tandem qui ont divergé et dont certaines sont maintenant utilisées de façon sélective pour le contournement de différents types de résistance. / Biological invasions in agroecosystems are a major cause of economic losses. Plant parasitic nematodes are among the many species causing significant crop damages. The soybean cyst nematode (SCN) is causing billions of dollars of losses in all areas where soybean is produced. Rotation with resistant cultivars is the most effective mean of controlling SCN populations, but the overuse of the same lines has led to the selection of virulent individuals and the ineffectiveness of resistance. To this day, the virulence genes and mecanisms associated with the circumvention of resistance continue to mystify scientists. In this thesis, I explored the effects of climate change on the reproduction and establishment of SCN as well as on the phenology of its host, soybean. I have demonstrated that the nematode can already reproduce in all regions of Québec and that the expected rise in temperatures in the near future (2041-2070) will allow the development of more generations per growing season in all regions. In addition, I have demonstrated that the area suitable for the production of soybean from maturity group I will expand toward the north by 2070, further facilitating the expansion of SCN. I have also explored the genetic variability among more than 64 SCN populations from North America and analyzed the genes associated with various bioclimatic components and their role in adaptation. These analyses revealed that the genetic diversity was very high among SCN populations. This diversity associated with a continuous gene flow between populations has facilitated the adaptation of SCN to various bioclimatic conditions and its establishment in all US and Canadian soybean producing regions. Finaly, this thesis presents an analysis of the SCN genotypes and the differentially expressed genes associated with virulence in two resistant soybean lines (Peking and PI88788) and susceptible Essex. This work has identified several proteins associated with virulence and allowed the discovery of a genomic region under strong evolutionary pressure. This island contains several genes in tandem duplications that have diverged and are now used selectively for overcoming different sources of resistance.

Prediction and understanding of the regulatory effects of non-coding DNA is an extensive research area in genomics. Convolutional neural networks have been used with success in the past to predict regulatory features, making chromatin feature predictions based solely on non-coding DNA sequences. Non-coding DNA shares various similarities with the human spoken language. This makes Language models such as the transformer attractive candidates for deciphering the non-coding DNA language. This thesis investigates how well the transformer model, usually used for NLP problems, predicts chromatin features based on genome sequences compared to convolutional neural networks. More specifically, the CNN DeepSEA, which is used for regulatory feature prediction based on noncoding DNA, is compared with the transformer DNABert. Further, this study explores the impact different parameters and training strategies have on performance. Furthermore, other models (DeeperDeepSEA and DanQ) are also compared on the same tasks to give a broader comparison value. Lastly, the same experiments are conducted on modified versions of the dataset where the labels cover different amounts of the DNA sequence. This could prove beneficial to the transformer model, which can understand and capture longrange dependencies in natural language problems. The replication of DeepSEA was successful and gave similar results to the original model. Experiments used for DeepSEA were also conducted on DNABert, DeeperDeepSEA, and DanQ. All the models were trained on different datasets, and their results were compared. Lastly, a Prediction voting mechanism was implemented, which gave better results than the models individually. The results showed that DeepSEA performed slightly better than DNABert, regarding AUC ROC. The Wilcoxon Signed-Rank Test showed that, even if the two models got similar AUC ROC scores, there is statistical significance between the distribution of predictions. This means that the models look at the dataset differently and might be why combining their prediction presents good results. Due to time restrictions of training the computationally heavy DNABert, the best hyper-parameters and training strategies for the model were not found, only improved. The Datasets used in this thesis were gravely unbalanced and is something that needs to be worked on in future projects. This project works as a good continuation for the paper Whole-genome deep-learning analysis identifies contribution of non-coding mutations to autism risk, Which uses the DeepSEA model to learn more about how specific mutations correlate with Autism Spectrum Disorder. / Arbetet kring hur icke-kodande DNA påverkar genreglering är ett betydande forskningsområde inom genomik. Convolutional neural networks (CNN) har tidigare framgångsrikt använts för att förutsäga reglerings-element baserade endast på icke-kodande DNA-sekvenser. Icke-kod DNA har ett flertal likheter med det mänskliga språket. Detta gör språkmodeller, som Transformers, till attraktiva kandidater för att dechiffrera det icke-kodande DNA-språket. Denna avhandling undersöker hur väl transformermodellen kan förutspå kromatin-funktioner baserat på gensekvenser jämfört med CNN. Mer specifikt jämförs CNN-modellen DeepSEA, som används för att förutsäga reglerande funktioner baserat på icke-kodande DNA, med transformern DNABert. Vidare undersöker denna studie vilken inverkan olika parametrar och träningsstrategier har på prestanda. Dessutom jämförs andra modeller (DeeperDeepSEA och DanQ) med samma experiment för att ge ett bredare jämförelsevärde. Slutligen utförs samma experiment på modifierade versioner av datamängden där etiketterna täcker olika mängder av DNA-sekvensen. Detta kan visa sig vara fördelaktigt för transformer modellen, som kan förstå beroenden med lång räckvidd i naturliga språkproblem. Replikeringen av DeepSEA experimenten var lyckad och gav liknande resultat som i den ursprungliga modellen. Experiment som användes för DeepSEA utfördes också på DNABert, DeeperDeepSEA och DanQ. Alla modeller tränades på olika datamängder, och resultat på samma datamängd jämfördes. Slutligen implementerades en algoritm som kombinerade utdatan av DeepDEA och DNABERT, vilket gav bättre resultat än modellerna individuellt. Resultaten visade att DeepSEA presterade något bättre än DNABert, med avseende på AUC ROC. Wilcoxon Signed-Rank Test visade att, även om de två modellerna fick liknande AUC ROC-poäng, så finns det en statistisk signifikans mellan fördelningen av deras förutsägelser. Det innebär att modellerna hanterar samma information på olika sätt och kan vara anledningen till att kombinationen av deras förutsägelser ger bra resultat. På grund av tidsbegränsningar för träning av det beräkningsmässigt tunga DNABert hittades inte de bästa hyper-parametrarna och träningsstrategierna för modellen, utan förbättrades bara. De datamängder som användes i denna avhandling var väldigt obalanserade, vilket måste hanteras i framtida projekt. Detta projekt fungerar som en bra fortsättning för projektet Whole-genome deep-learning analysis identifies contribution of non-coding mutations to autism risk, som använder DeepSEA-modellen för att lära sig mer om hur specifika DNA-mutationer korrelerar med autismspektrumstörning.

Whole-exome and genome sequencing are widely used to diagnose individual patients. However, despite its success, this approach leaves many patients undiagnosed. This could be due to the need to discover more disease genes and variants or because disease phenotypes are novel and arise from a combination of variants of multiple known genes related to the disease. Recent rapid increases in available genomic, biomedical, and phenotypic data enable computational analyses, reducing the search space for disease-causing genes or variants and facilitating the prediction of causal variants. Therefore, artificial intelligence, data mining, machine learning, and deep learning are essential tools that have been used to identify biological interactions, including protein-protein interactions, gene-disease predictions, and variant--disease associations. Predicting these biological associations is a critical step in diagnosing patients with rare or complex diseases. In recent years, computational methods have emerged to improve gene-disease prioritization by incorporating phenotype information. These methods evaluate a patient's phenotype against a database of gene-phenotype associations to identify the closest match. However, inadequate knowledge of phenotypes linked with specific genes in humans and model organisms limits the effectiveness of the prediction. Information about gene product functions and anatomical locations of gene expression is accessible for many genes and can be associated with phenotypes through ontologies and machine-learning models. Incorporating this information can enhance gene-disease prioritization methods and more accurately identify potential disease-causing genes. This dissertation aims to address key limitations in gene-disease prediction and variant prioritization by developing computational methods that systematically relate human phenotypes that arise as a consequence of the loss or change of gene function to gene functions and anatomical and cellular locations of activity. To achieve this objective, this work focuses on crucial problems in the causative variant prioritization pipeline and presents novel computational methods that significantly improve prediction performance by leveraging large background knowledge data and integrating multiple techniques. Therefore, this dissertation presents novel approaches that utilize graph-based machine-learning techniques to leverage biomedical ontologies and linked biological data as background knowledge graphs. The methods employ representation learning with knowledge graphs and introduce generic models that address computational problems in gene-disease associations and variant prioritization. I demonstrate that my approach is capable of compensating for incomplete information in public databases and efficiently integrating with other biomedical data for similar prediction tasks. Moreover, my methods outperform other relevant approaches that rely on manually crafted features and laborious pre-processing. I systematically evaluate our methods and illustrate their potential applications for data analytics in biomedicine. Finally, I demonstrate how our prediction tools can be used in the clinic to assist geneticists in decision-making. In summary, this dissertation contributes to the development of more effective methods for predicting disease-causing variants and advancing precision medicine.

Genregulation beeinflusst Phänotypen im Kontext von Gesundheit und Krankheit. In Krebszellen regulieren genetische und epigenetische Faktoren die Genexpression in cis. Das Neuroblastom ist eine Krebserkrankung, die häufig im Kindesalter auftritt. Es ist gekennzeichnet durch eine geringe Anzahl exonischer Mutationen und durch häufige Veränderungen der somatischen Kopienzahl, einschließlich Genamplifikationen auf extrachromosomaler zirkulärer DNA. Bisher ist wenig darüber bekannt, wie lokale genetische und epigenetische Faktoren Gene im Neuroblastom regulieren. In dieser Arbeit kombiniere ich die allelspezifische Analyse ganzer Genome (WGS), Transkriptome und zirkulärer DNA von Neuroblastom-Patienten, um genetische und cis-regulatorische Effekte zu charakterisieren. Ich zeige, dass somatische Dosis-Effekte der Kopienzahl andere lokale genetische Effekte dominieren und wichtige Signalwege regulieren. Genamplifikationen zeigen starke Dosis-Effekte und befinden sich häufig auf großen extrachromosomalen zirkulären DNAs. Die vorgestellte Analyse zeigt, dass der Verlust von 11q zu einer Hochregulation von Histonvarianten H3.3 und H2A in Tumoren mit alternativer Verlängerung der Telomere (ALT) führt, und dass erhöhte somatische Kopienzahl die Expression der TERT Gens verstärken können. Weitere Erkenntnisse sind, dass 17p-Ungleichgewichte und die damit verbundene Herunterregulierung neuronaler Gene sowie die Hochregulierung des genomisch geprägten Gens RTL1 durch Kopienzahl-unabhängige allelische Dosis-Effekte mit einer ungünstigen Prognose verbunden sind. Die cis-QTL-Analyse bestätigt eine zuvor beschriebene Regulation des LMO1 Gens durch einen Enhancer-Polymorphismus und charakterisiert das regulatorische Potenzial weiterer GWAS-Risiko-Loci. Die Arbeit unterstreicht die Bedeutung von Dosis-Effekten im Neuroblastom und liefert eine detaillierte Übersicht regulatorischer Varianten, die in dieser Krankheit aktiv sind. / Gene regulation controls phenotypes in health and disease. In cancer, the interplay between germline variation, genetic aberrations and epigenetic factors modulate gene expression in cis. The childhood cancer neuroblastoma originates from progenitor cells of the sympathetic nervous system. It is characterized by a sparsity of recurrent exonic mutations but frequent somatic copy-number alterations, including gene amplifications on extrachromosomal circular DNA. So far, little is known on how local genetic and epigenetic factors regulate genes in neuroblastoma to establish disease phenotypes. I here combine allele-specific analysis of whole genomes, transcriptomes and circular DNA from neuroblastoma patients to characterize genetic and cis-regulatory effects, and prioritize germline regulatory variants by cis-QTLs mapping and chromatin profiles. The results show that somatic copy-number dosage dominates local genetic effects and regulates pathways involved in telomere maintenance, genomic stability and neuronal processes. Gene amplifications show strong dosage effects and are frequently located on large but not small extrachromosomal circular DNAs. My analysis implicates 11q loss in the upregulation of histone variants H3.3 and H2A in tumors with alternative lengthening of telomeres and cooperative effects of somatic rearrangements and somatic copy-number gains in the upregulation of TERT. Both 17p copy-number imbalances and associated downregulation of neuronal genes as well as upregulation of the imprinted gene RTL1 by copy-number-independent allelic dosage effects is associated with an unfavorable prognosis. cis-QTL analysis confirms the previously reported regulation of the LMO1 gene by a super-enhancer risk polymorphism and characterizes the regulatory potential of additional GWAS risk loci. My work highlights the importance of dosage effects in neuroblastoma and provides a detailed map of regulatory variation active in this disease.