Most medical treatments have been developed aiming at the best-on-average efficacy for large populations, resulting in treatments successful for some patients but not for others. It necessitates the need for precision medicine that tailors medical treatment to individual patients. Omics data holds comprehensive genetic information on individual variability at the molecular level and hence the potential to be translated into personalized therapy. However, the attempts to transform omics data-driven insights into clinically actionable models for individual patients have been limited. Meanwhile, advances in deep learning, one of the most promising branches of artificial intelligence, have produced unprecedented performance in various fields. Although several deep learning-based methods have been proposed to predict individual phenotypes, they have not established the state of the practice, due to instability of selected or learned features derived from extremely high dimensional data with low sample sizes, which often results in overfitted models with high variance. To overcome the limitation of omics data, recent advances in deep learning models, including representation learning models, generative models, and interpretable models, can be considered. The goal of the proposed work is to develop deep learning models that can overcome the limitation of omics data to enhance the prediction of personalized medical decisions. To achieve this, three key challenges should be addressed: 1) effectively reducing dimensions of omics data, 2) systematically augmenting omics data, and 3) improving the interpretability of omics data. / Doctor of Philosophy / Most medical treatments have been developed aiming at the best-on-average efficacy for large populations, resulting in treatments successful for some patients but not for others. It necessitates the need for precision medicine that tailors medical treatment to individual patients. Biological data such as DNA sequences and snapshots of genetic activities hold comprehensive information on individual variability and hence the potential to accelerate personalized therapy. However, the attempts to transform data-driven insights into clinical models for individual patients have been limited. Meanwhile, advances in deep learning, one of the most promising branches of artificial intelligence, have produced unprecedented performance in various fields. Although several deep learning-based methods have been proposed to predict individual treatment or outcome, they have not established the state of the practice, due to the complexity of biological data and limited availability, which often result in overfitted models that may work on training data but not on test data or unseen data. To overcome the limitation of biological data, recent advances in deep learning models, including representation learning models, generative models, and interpretable models, can be considered. The goal of the proposed work is to develop deep learning models that can overcome the limitation of omics data to enhance the prediction of personalized medical decisions. To achieve this, three key challenges should be addressed: 1) effectively reducing the complexity of biological data, 2) generating realistic biological data, and 3) improving the interpretability of biological data.

No / Skeletal muscle has long been established as a highly multifunctional organ, playing a vital role in locomotion, whole-body metabolic and energy homeostasis, and thermoregulation. More recently, emergent evidence has highlighted a potent secretory role for muscle, producing and releasing “myokine” molecules that act in autocrine, paracrine, or endocrine fashion to govern muscle physiology and regulate whole-body homeostasis via multi-tissue cross talk mechanisms. Myokines represent promising therapeutic targets in health and disease, with their discovery, measurement, and functional importance being a hotbed of research across numerous physiological contexts. Here, we provide an overview of myokines and summarize current understanding of their biological role(s). We also outline primary approaches for myokine analysis, including detailed methodology for performing omics-driven myokine prediction, while further appraising both method-specific and general technical considerations to provide an evidence-based approach for designing and conducting myokine experiments.

L'ostéochondrose (OC) est une affection ostéo-articulaire juvénile caractérisée par une perturbation locale de la maturation du cartilage créant des zones de fragilité. L'OC a été décrite chez de nombreuses espèces dont l'Homme, le Chien, le Porc, le Poulet et le Cheval. Chez le cheval les lésions s'installent progressivement, sans symptômes, avant l'âge d'un an et les manifestations cliniques ne se manifestent que tardivement, souvent à l'entraînement. L'OC affecte 10 à 30 % de la population équine représentant ainsi un souci majeur pour la filière tant sur le plan du bien être animal que sur le plan économique. Son étiologie, multifactorielle, est encore mal comprise et implique des composantes génétiques et environnementales ainsi que traumatiques. Les objectifs des travaux présentés étaient d'améliorer la compréhension de la physiopathologie de l'OC équine et de mettre en évidence les processus biologiques perturbés. L'ensemble des résultats a permis de préciser la définition des différentes entités de l'OC et pourraient également être pertinentes dans l'amélioration du diagnostic et le dévelopement de nouveaux traitements. Un défaut constitutif de l'os et du cartilage a été mis en évidence chez les individus atteints d'OC, notamment associé à une perturbation du métabolisme énergénique et un stress du reticulum endoplasmique. De plus, selon le type de lésions, des mécanismes moléculaires sous-jacents différents sont impliqués dans leur développement. D'autre part, les microARNs (miRNAs) semblent également jouer un rôle dans la physiopathologie de l'OC et certains d'entre eux pourraient constituer de bonnes cibles thérapeutiques ou être utilisés comme biomarqueurs diagnostics. / Osteochondrosis (OC) is a juvenile osteo-articular disease characterized by a focal failure of cartilage maturation leading to weak areas. OC has been described in several species including Human, Dog, Swine, Poultry and Horse. In horse, lesions develop gradually without symptoms before one year old and clinical manifestations occur tardily during training. OC affects 10 to 30% of equine population and constitutes a major concern in terms of animal welfare and economy. Its multifactorial etiology remains poorly understood and involved several factors including genetics, environment and traumas. The aim of this current work was to improve the comprehension of equine OC physiopathology and highlight biological pathways disrupted. Taken together, our results made it possible to refine the definition of OC entities and our data could be relevant to improve diagnosis and develop new therapies. A constitutive defect was found in cartilage and bone of OC-affected horses and particularly a defective energy metabolism and a endoplasmic reticulum stress. Moreover, in function of lesion type, different underlying molecular mechanisms are involved in their development. Secondly, mircoRNAs (miRNAs) seem to take part in the OC physiopathology and some miRNAs could constitute a relevant therapeutic target or be used as diagnosis biomarkers.

One of the greatest risk factors for disease is advanced age. As the human lifespan has increased, so too have the burdens of caring for an increasingly older population suffering from rising rates of cardiovascular disease, kidney disease, diabetes, and dementia. The need for improving medical technology and developing new therapies for age-related diseases is manifest. Yet our understanding of the processes of aging and how to attenuate the effects of aging remains incomplete. Various studies have established calorie restriction as a robust method for extending lifespan in laboratory organisms; however the mechanism is a topic of much debate. Advancing our understanding of calorie restriction holds promise for illuminating biochemical processes involved in the aging process. One of the best explanations for the lifespan extension benefits of calorie restriction is that it improves cellular protein homeostasis (proteostasis), but because proteostasis is dynamic, it can be difficult to measure. We developed a novel combined omics methodology integrating kinetic proteomics, and applied it to a mouse model placed on calorie restriction. Our unbiased approach integrating just three measurements (kinetic proteomics, quantitative proteomics, and transcriptomics) enabled us to characterize the synthesis and degradation of thousands of proteins, and determine that calorie restriction largely alters proteostasis by slowing global protein synthesis post-transcriptionally. Validating our omics approach, we were able to replicate many previous results found in the literature, demonstrating the differential regulation of various protein ontologies in response to the nutrient stress of calorie restriction. Moreover, we were able to detect differential degradation of the large and small ribosomal subunits under calorie restriction, and proposed a model in which the rate of protein synthesis could be attenuated by the depletion of the large ribosomal subunit relative to the small subunit. The flexibility of our dynamic combined omics approach was demonstrated by the expansion of measurements to include nucleic acids and lipids. Flux measurements of DNA, ribosomal RNA, and lipids yielded cellular division rates, ribosome turnover, and lipid metabolism insights, respectively. We also adapted this approach to two-dimensional tissue imaging by DESI-MS in a proof-of-concept study to demonstrate its utility for studying regional differences in metabolism. The future integration of metabolomics and lipidomics into our combined omics approach would be facile, and add unprecedented depth to systems-wide studies involving cellular metabolism. Applied to the regulation of cellular homeostasis in humans, this has the potential to open new avenues for elucidating the etiology of aging, understanding the pathology of age-related diseases, and identifying novel targets for therapeutics.

Les efflorescences de cyanobactéries sont susceptibles d’avoir des effets néfastes sur les organismes des écosystèmes aquatiques, ainsi que sur les populations environnantes, notamment à travers la production de nombreuses molécules potentiellement toxiques (appelées cyanotoxines). Jusqu'à présent, une des cyanotoxines les plus étudiées est la microcystine (MC). Cette thèse a pour objectif d’évaluer la toxicité potentielle sur la reproduction de la MC-LR et de l'extrait d'une souche de Microcystis productrice de MCs en étudiant leurs effets toxiques sur le foie et les gonades de poissons medaka adultes exposés de manière aiguë ou chronique.Une étude complète du foie des poissons médaka deux sexes a été menés par ailleurs, attestant d'un fort dimorphisme sexuel aussi bien au niveau cellulaire que moléculaire et souligne les importantes spécificités métaboliques du foie entre les deux sexes, notamment pour le maintien de la compétence de reproduction chez les poissons medaka adultes femelles.Dans l'étude des effets induits par une exposition aiguë, les poissons medaka adultes ont été exposés par gavage à 10 μg.g-1 bw de MC-LR pure pendant 1 heure. L'examen histologique et l'immunolocalisation des MCs du foie de poisson traité par la MC-LR ont révélé des lésions hépatiques sévères ainsi qu'une distribution intense de la MC-LR dans le foie, localisée particulièrement dans le cytoplasme et dans le noyau des hépatocytes. Dans la gonade des poissons traités, la MC-LR a été détectée dans les tissus conjonctifs de l'ovaire et des testicules. De plus, l’observation par microscopie électronique couplé à la technique d’immunogold a révélé, pour la première fois, que la MC-LR était également détectable dans le chorion, le cytoplasme et le vitellus des ovocytes matures.Au cours des études des effets induits par l’exposition chronique, les poissons medaka adultes ont été exposés durant 28 jours par balnéation à 1 et 5 μg.L-1 de MC-LR et à un extrait de la souche de Microcystis aeruginosa (PCC 7820) productrice de microcystines (5 μg.L-1 équivalent MC-LR). Ces résultats ont révélé que la MC-LR et l'extrait de Microcystis induisent des effets délétères sur différents paramètres de reproduction, tels la fécondité et le taux d’éclosion des embryons. La cause principale de ces perturbations de la reproduction semblent principalement résulté d’un dysfonction hépatique globale induite par les traitements aux MCs (hépatotoxiques, notoires), plutôt qu’à des effets directs sur les gonades. Dans l'ensemble, les résultats de cette thèse démontrent que même si les microcystines pourraient avoir un impact direct, mais modéré, sur la fonction gonadique en induisant une cytotoxicité dans les cellules somatiques gonadiques et les cellules reproductrices, elle semble avoir principalement avoir un impact indirect sur la fonction reproductrice en perturbant la fonction hépatique générale. Ces données améliorent notre compréhension des processus liés à la toxicité potentielle des cyanotoxines pour la reproduction chez un poisson modèle, et fait d’une manière générale progresser questionnement quant à la protection des populations exposées à ces cyanotoxines. / Cyanobacterial blooms threaten human health as well as other living organisms of the aquatic environment, particularly due to the production of natural toxic components (called cyanotoxins). So far, one of the most studied cyanotoxins is the microcystin (MC). This thesis evaluated the potential reproductive toxicity of MC-LR and the extract of one Microcystis strain (MC-producing) by investigating their toxic effects on the liver and gonad of adult medaka fish with one acute and one chronic study.An investigation of the metabolic specificities of the liver in two genders of medaka fish was performed prior to the MC-containing exposure, which attests to a strong sexual dimorphism of medaka liver, and highlights the importance of metabolic adjustments of the liver for maintaining the reproductive competency in adult medaka fish.In the acute study, adult medaka fish were administered with 10 μg.g-1 bw of pure MC-LR for 1 hour by gavage. The histological examination and immunolocalization of the MC-treated fish liver revealed a severe liver lesion along with an intense distribution of MC-LR in the liver, being particularly localized in the cytoplasm and nucleus of hepatocytes. In the gonad of MC-treated fish, MC-LR was shown to be present in the connective tissue of ovary and testis. Additionally, immunogold electron microscopy, for the first time, revealed that MC-LR was also localized in the chorion, cytoplasm and yolk vesicles of oocytes.Overall, the results of this thesis demonstrates that MC might directly impact gonadal function by inducing cytotoxicity in gonadal somatic cells and reproductive cells, and it could also impact the reproductive function indirectly by disturbing the general liver function. This improves our understanding of the potential reproductive toxicity of cyanotoxins in model fish, and advances our current knowledge on the protection of aquatic organism populations as well as human health from cyanotoxin issues.

Most research on human pancreatic islets is conducted on samples obtained from normoglycaemic or diseased brain-dead donors and thus cannot accurately describe the molecular changes of pancreatic islet beta cells as they progress towards a state of deficient insulin secretion in type 2 diabetes (T2D). Here, we conduct a comprehensive multi-omics analysis of pancreatic islets obtained from metabolically profiled pancreatectomized living human donors stratified along the glycemic continuum, from normoglycemia to T2D. We find that islet pools isolated from surgical samples by laser-capture microdissection display remarkably more heterogeneous transcriptomic and proteomic profiles in patients with diabetes than in non-diabetic controls. The differential regulation of islet gene expression is already observed in prediabetic individuals with impaired glucose tolerance. Our findings demonstrate a progressive, but disharmonic, remodelling of mature beta cells, challenging current hypotheses of linear trajectories toward precursor or transdifferentiation stages in T2D. Furthermore, through integration of islet transcriptomics with preoperative blood plasma lipidomics, we define the relative importance of gene coexpression modules and lipids that are positively or negatively associated with HbA1c levels, pointing to potential prognostic markers.

[ES] El objetivo de la Tesis fue investigar los mecanismos genéticamente determinados involucrados en la deposición de grasa intramuscular (GIM) usando 2 líneas divergentes de GIM (A y B) en el músculo Longissimus thoracis et Lumborum (LTL) de conejos. Consta de cinco estudios que evalúan el rol del huésped y el microbioma usando análisis metabolómicos y metagenómicos. La respuesta a la selección en la 10ma generación fue 0,49g GIM/100g LTL, equivalente a 3,8 desviaciones estándar (DE). Se obtuvo una respuesta correlacionada positiva en la grasa de la canal y cambios en los ácidos grasos (AG) del LTL, con mayor contenido de saturados (A-B= 5,05g/100g GIM) y monoinsaturados (A-B= 5,04g/100g GIM) en la línea A. No hubo diferencia en la grasa del hígado, pero sí en sus AG, como el menor C15:0 (A-B= -0,04g/100g lípidos) y C17:0 (A-B= -0,09g/100g lípidos) en la línea A, que podría deberse a diferente digestión microbiana. El análisis metabolómico del plasma encontró 393 metabolitos diferenciales con 95% precisión clasificatoria y 383 metabolitos con ajuste linear a GIM con 65% capacidad predictiva, de los cuales 322 coincidían con diferencias entre -6,04 y +1,97 DE. Los lípidos fueron mayores en la línea B (ej. triglicéridos, ácidos biliares secundarios (AB-2º), AG) y la carnitina fue menor, sugiriendo mayor absorción intestinal y menor captación y almacenamiento, posiblemente relacionada con menor ß-oxidación de AG. Entre los aminoácidos, destacaron los de cadena ramificada (BCAA) y aromáticos (AAA) indicando menor degradación intestinal de BCAA en la línea A seguida de mayor catabolismo del huésped, y una compleja interacción huésped-microbioma en el metabolismo de los AAA. El análisis metagenómico del ciego confirmó la relevancia del microbioma en la deposición de GIM, con cambios en composición y funcionalidad. El análisis de la composición definió 2 enterotipos con 51 géneros microbianos y 91% precisión clasificatoria: el enterotipo A enriquecido en Hungateiclostridium, Limosilactobacillus, Legionella, Lysinibacillus, Phorphyromonas, Methanosphaera y Desulfovibrio y el enterotipo B en Escherichia, Fonticella, Candidatus Amulumruptor, Methanobrevicater, Exiguobacterium, Flintibacter y Coprococcus. Un balance composicional se propuso como biomarcador para predecir la predisposición genética a la deposición de GIM, compuesto por 26 géneros microbianos con 93% precisión clasificatoria y 69% capacidad predictiva. El análisis de funcionalidad encontró 240 genes microbianos (GM) diferenciales con 95% precisión clasificatoria y 230 GM con ajuste linear a GIM con 79% capacidad predictiva, de los cuales 122 GM coincidían con diferencias entre -0,75 y +0,73 DE. Mayor biosíntesis de lipopolisacáridos y peptidoglicanos, asociado al desarrollo de masa grasa, biosíntesis de AAA, asociado a trastornos relacionados a la deposición grasa, y conversión de propionato a acetato, relacionado con mayor lipogénesis en el hígado y menor síntesis de C15:0 y C17:0, se encontró en la línea A. Además, se encontró mayor degradación de BCAA en la línea B, asociado con menor síntesis de triglicéridos en el hígado. El análisis metabolómico del ciego encontró 142 metabolitos diferenciales con 99% precisión clasificatoria y diferencias entre -1,03 y +1,19 DE; 156 relacionados con GIM en la línea A con 61% capacidad predictiva; y 107 relacionados con GIM en la línea B con 57% capacidad predictiva. Diferencias en el metabolismo de las purinas podrían sugerir mayor eficiencia energética y en la utilización del nitrógeno en la línea B, y diferencias en AB-2º, AAA y BCAA fueron consistentes con los resultados anteriores. Un balance composicional se propuso como biomarcador compuesto por 2 AB-2º y 2 subproductos de las proteínas, con 88% de precisión clasificatoria, sugiriendo que la interacción entre absorción lipídica y metabolismo de proteínas de la dieta influyen en la GIM. De validarse, podría usarse para predecir la predisposición genética a la deposición de GIM. / [CA] L'objectiu de la Tesi va ser investigar els mecanismes genèticament determinats involucrats en la deposició de greix intramuscular (GIM) usant 2 línies divergents de GIM (A i B) en el múscul Longissimus thoracis et Lumborum (LTL) de conills. Consta de cinc estudis que avaluen el rol de l'hoste i el microbioma usant anàlisi metabolòmicos i metagenòmicos. La resposta a la selecció en la 10ma generació va ser 0,49g GIM/100g LTL, equivalent a 3,8 desviacions estàndard (DE) . Es va obtindre una resposta correlacionada positiva en el greix de la canal i canvis en els àcids grassos (AG) del LTL, amb major contingut de saturats (A-B = 5,05g/100g GIM) i monoinsaturats (A-B = 5,04g/100g GIM) en la línia A. No va haver-hi diferència en el greix del fetge, però sí en els seus AG, com el menor C15:0 (A-B = -0,04g/100g lípids) i C17:0 (A-B = -0,09g/100g lípids) en la línia A, que podria deure's a diferent digestió microbiana. L'anàlisi metabolómico del plasma va trobar 393 metabòlits diferencials amb 95% precisió classificatòria i 383 metabòlits amb ajust linear a GIM amb 65% capacitat predictiva, dels quals 322 coincidien amb diferències entre -6,04 i 1,97 DE. Els lípids van ser majors en la línia B (ex. triglicèrids, àcids biliars secundaris (AB- 2º), AG) i la carnitina va ser menor, suggerint major absorció intestinal i menor captació i emmagatzematge, possiblement relacionada amb menor ß-oxidació d'AG. Entre els aminoàcids, van destacar els de cadena ramificada (BCAA) i aromàtics (AAA) indicant menor degradació intestinal de BCAA en la línia A seguida de major catabolisme de l'hoste, i una complexa interacció hoste-microbioma en el metabolisme dels AAA. L'anàlisi metagenòmico del cec va confirmar la rellevància del microbioma en la deposició de GIM, amb canvis en composició i funcionalitat. L'anàlisi de la composició va definir 2 enterotipos amb 51 gèneres microbians i 91% precisió classificatòria: el enterotipo A enriquit en Hungateiclostridium, Limosilactobacillus, Legionel·la, Lysinibacillus, Phorphyromonas, Methanosphaera i Desulfovibrio i el enterotipo B en Escherichia, Fonticella, Candidatus Amulumruptor, Methanobrevicater, Exiguobacterium, Flintibacter i Coprococcus. Un balanç composicional es va proposar com biomarcador per a predir la predisposició genètica a la deposició de GIM, compost per 26 gèneres microbians amb 93% precisió classificatòria i 69% capacitat predictiva. L'anàlisi de funcionalitat va trobar 240 gens microbians (GM) diferencials amb 95% precisió classificatòria i 230 GM amb ajust linear a GIM amb 79% capacitat predictiva, dels quals 122 GM coincidien amb diferències entre -0,75 i 0,73 DE. Major biosíntesi de lipopolisacàrids i peptidoglicans, associat al desenvolupament de massa grassa, biosíntesi de AAA, associat a trastorns relacionats a la deposició grassa, i conversió de propionat a acetat, relacionat amb major lipogènesis en el fetge i menor síntesi de C15:0 i C17:0, es va trobar en la línia A. A més, es va trobar major degradació de BCAA en la línia B, associat amb menor síntesi de triglicèrids en el fetge. L'anàlisi metabolòmico del cec va trobar 142 metabòlits diferencials amb 99% precisió classificatòria i diferències entre -1,03 i 1,19 DE; 156 relacionats amb GIM en la línia A amb 61% capacitat predictiva; i 107 relacionats amb GIM en la línia B amb 57% capacitat predictiva. Diferències en el metabolisme de les purins podrien suggerir major eficiència energètica i en la utilització del nitrogen en la línia B, i diferències en AB-2º, AAA i BCAA van ser consistents amb els resultats anteriors. Un balanç composicional es va proposar com biomarcador compost per 2 AB-2º i 2 subproductes de les proteïnes, amb 88% de precisió classificatòria, suggerint que la interacció entre absorció lipídica i metabolisme de proteïnes de la dieta influeixen en la GIM. De validar-se, podria usar-se per a predir la predisposició genètica a la deposició de GIM. / [EN] The Thesis aimed to study the genetically determined mechanisms involved in intramuscular fat (IMF) deposition, using two lines divergently selected for IMF in Longissimus thoracis et Lumborum (LTL) muscle of rabbits (H and L lines). It comprises five studies focused on studying the host and microbiome roles using metabolomics and metagenomics approaches. The response to selection in the 10th generation was 0.49g IMF/100g LTL, equivalent to 3.8 standard deviations (SD). Selection led to a positive correlated response in carcass adiposity, and to changes in the fatty acids (FA) of LTL, showing greater saturated (H-L= 5.05g/100g IMF) and monounsaturated FA (H-L= 5.04g/100g IMF) in the H line. No differences were found in liver fat, but they were found in its FA profile, being the most notorious the lower C15:0 (H-L= -0.04g/100g lipids) and C17:0 (H-L= -0.09g/100g lipids) in the H line, which could be due to different microbial digestion. The plasma metabolomics analysis identified 393 differential metabolites with 95% classification accuracy, and 383 metabolites with linear adjustment to IMF and 65% prediction ability, from which 322 overlapped with differences ranging from -6.04 to +1.97 SD. Lipids were greater in the L line (e.g., triglycerides, secondary bile acids, FA) while carnitine was lower, suggesting greater intestinal lipids absorption in the L line, followed by their lower uptake and storage, possibly related to lower FA ß oxidation. Among amino acids, branched-chain (BCAA) and aromatic (AAA) stood out, indicating lower BCAA gut degradation in the H line followed by greater host catabolism, and a complex host-microbiome AAA metabolism. The caecum metagenomics analysis confirmed the microbial relevance in IMF development, identifying changes in its composition and functionality. The microbial composition analysis defined two enterotypes with 51 microbial genera and 91% classification accuracy. The H-enterotype was enriched in Hungateiclostridium, Limosilactobacillus, Legionella, Lysinibacillus, Phorphyromonas, Methanosphaera and Desulfovibrio, and the L-enterotype in Escherichia, Fonticella, Candidatus Amulumruptor, Methanobrevicater, Exiguobacterium, Flintibacter and Coprococcus. A compositional balance was proposed as biomarker to predict the genetic predisposition to IMF deposition, composed of 26 microbial genera, with 93% classification accuracy and 69% prediction ability. The microbial functionality analysis identified 240 differential microbial genes (MG) with 95% classification accuracy, and 230 MG with linear adjustment to IMF and 79% prediction ability, from which 122 overlapped with differences ranging from -0.75 to +0.73 SD and related to numerous metabolisms. In the H line, greater lipopolysaccharides and peptidoglycans biosynthesis, related to fat-mass development, AAA biosynthesis, related to associated disorders of increased fat deposition, and propionate to acetate conversion, related to greater liver lipogenesis and lower C15:0 and C17:0 synthesis, were found. Additionally, greater BCAA degradation was found in the L line, related to lower triglycerides synthesis in the liver. The caecum metabolomics analysis identified 142 differential metabolites with 99% classification accuracy and differences ranging from -1.03 to +1.19 SD; 156 related to IMF in the H line with 61% prediction ability; and 107 related to IMF in the L line with 57% prediction ability. Differences found in purine metabolism could suggest greater energy and nitrogen utilization efficiencies in the L line, while those in secondary bile acids, AAA and BCAA were consistent with the previous results. A compositional balance composed of two secondary bile acids and two proteins by-products with 88% classification accuracy was proposed as biomarker, suggesting that the interaction between lipids absorption and dietary proteins metabolism influences IMF. If validated, it could be used to predict the genetic predisposition to IMF deposition. / Zubiri Gaitán, A. (2024). Intramuscular Fat Deposition in Rabbits: Insights into Host-Microbiome Biological Mechanisms [Tesis doctoral]. Universitat Politècnica de València. https://doi.org/10.4995/Thesis/10251/202875

Thesis (PhD)--Stellenbosch University, 2013. / ENGLISH ABSTRACT: Introduction: Wine production is a complex biotechnological process aiming at productively coordinating the interactions and outputs of several biological systems, including grapevine and many microorganisms such as wine yeast and wine bacteria. High-throughput data generating tools in the elds of genomics, transcriptomics, proteomics, metabolomics and microbiomics are being applied both locally and globally in order to better understand complex biological systems. As such, the datasets available for analysis and mining include de novo datasets created by collaborators as well as publicly available datasets which one can use to get further insight into the systems under study. In order to model the complexity inherent in and across these datasets it is necessary to develop methods and approaches based on network theory and multivariate data analysis as well as to explore the intersections between these two approaches to data modelling, mining and interpretation. Networks: The traditional reductionist paradigm of analysing single components of a biological system has not provided tools with which to adequately analyse data sets that are attempting to capture systems-level information. Network theory has recently emerged as a new discipline with which to model and analyse complex systems and has arisen from the study of real and often quite large networks derived empirically from the large volumes of data that have collected from communications, internet, nancial and biological systems. This is in stark contrast to previous theoretical approaches to understanding complex systems such as complexity theory, synergetics, chaos theory, self-organised criticality, and fractals which were all sweeping theoretical constructs based on small toy models which proved unable to address the complexity of real world systems. Multivariate Data Analysis: Principle components analysis (PCA) and Partial Least Squares (PLS) regression are commonly used to reduce the dimensionality of a matrix (and amongst matrices in the case of PLS) in which there are a considerable number of potentially related variables. PCA and PLS are variance focused approaches where components are ranked by the amount of variance they each explain. Components are, by de nition, orthogonal to one another and as such, uncorrelated. Aims: This thesis explores the development of Computational Biology tools that are essential to fully exploit the large data sets that are being generated by systems-based approaches in order to gain a better understanding of winerelated organisms such as grapevine (and tobacco as a laboratory-based plant model), plant pathogens, microbes and their interactions. The broad aim of this thesis is therefore to develop computational methods that can be used in an integrated systems-based approach to model and describe di erent aspects of the wine making process from a biological perspective. To achieve this aim, computational methods have been developed and applied in the areas of transcriptomics, phylogenomics, chemiomics and microbiomics. Summary: The primary approaches taken in this thesis have been the use of networks and multivariate data analysis methods to analyse highly dimensional data sets. Furthermore, several of the approaches have started to explore the intersection between networks and multivariate data analysis. This would seem to be a logical progression as both networks and multivariate data analysis are focused on matrix-based data modelling and therefore have many of their roots in linear algebra. / AFRIKAANSE OPSOMMING: Inleiding: Wynproduksie is 'n komplekse biotegnologiese proses wat mik op die produktiewe koördinering van verskeie interaksies en uitsette van verskeie biologiese sisteme. Hierdie sisteme sluit in die wingerd, wat van besondere belang is, asook die wyn gis en wyn bakterieë. Hoë-deurset data generasie word huidiglik beide globaal en plaaslik toegepas in die velde van genomika, transkriptomika, proteomika, metabolomika en mikrobiomika. As sulks is hierdie tipe datastelle beskikbaar vir ontleding, bemyning en verkening. Die datastelle kan de novo gegenereer word, met behulp van medewerkers, of dit kan vanuit die publieke databasisse gewerf word waar sulke datastelle dikwels beskikbaar gemaak word sodat verdere insig verkry kan word met betrekking tot die sisteem onder studie. Die hoë-deurset datastelle onder bespreking bevat 'n hoë mate van inherente kompleksiteit, beide ten opsigte van ditself asook tussen verskeie datastelle. Om ten einde hierdie datastelle en hul inherente kompleksiteit te modelleer is dit nodig om metodes en benaderings te ontwikkel wat gesetel is in netwerk teorie en meerveranderlike statistiek. Verdermeer is dit ook nodig om die kruisings tussen netwerk teorie en meerveranderlike statistiek te verken om sodoende die modellering, bemyning, verkening en interpretasie van data te verbeter. Netwerke: Die tradisionele reduksionistiese paradigma, waarby enkele komponente van 'n biologiese sisteem geontleed word, het tot dusver nie voldoende metodes en gereedskap gelewer waarmee datastelle, wat streef om sisteemvlak informasie te bekom, geontleed kan word nie. Netwerk teorie het na vore gekom as 'n nuwe dissipline wat toegepas kan word vir die model-skepping en ontleding van komplekse sisteme. Dit stem uit die studie van egte, dikwels groot netwerke wat empiries afgelei word uit die groot volumes data wat tans na vore kom vanuit kommunikasie-, internet-, nansiële- en biologiese sisteme. Dit is in skrille kontras met vorige teoretiese benaderings wat gestreef het om komplekse sisteme te verstaan met konsepte soos kompleksiteits teorie, synergetics , chaos teorie, self-georganiseerde kritikaliteit en fraktale. Al die bogeneomde is breë teoretiese konstrukte, gebasseer op relatief kleinskaal modelle, wat nie instaat was om oplossings vir die kompleksiteit van egte-wêreld sisteme te bied nie. Meerveranderlike Data-analise: Hoofkomponente-ontleding (PCA) en Partial Least Squares (PLS) regressie word dikwels gebruik om die dimensionaliteit van 'n matriks (en tussen matrikse in die geval van PLS) te verminder. Hierdie matrikse bevat dikwels 'n aansienlike groot hoeveelheid moontlikverwante veranderlikes. PCA en PLS is variansie gedrewe metodes en behels dat komponente gerang word deur die hoeveelheid variansie wat elke component verduidelik. Komponente is by de nisie ortogonaal ten opsigte van mekaar en as sulks ongekorreleerd. Doelwitte: Hierdie tesis verken die ontwikkeling van verskeie Computational Biology metodes wat noodsaaklik is om ten volle die groot skaal datastelle te benut wat tans deur sisteem-gebasseerde benaderings gegenereer word. Die doel is om beter begrip en kennis van wyn verwante organismes te kry, hierdie organismes sluit in die wingerd (met tabak as laboratorium-gebasseerde plant model), plant patogene en microbes sowel as hulle interaksies. Die breë mikpunt van hierdie tesis is dus om gerekenaardiseerde metodes te ontwikkel wat gebruik kan word in 'n geintergreerde sisteem-gebaseerde benadering tot die modellering en beskrywing van verskillende aspekte van die wynmaak proses vanuit 'n biologiese standpunt. Om die mikpunt te bereik is gerekenaardiseerde metodes ontwikkel en toegepas in die velde van transkriptomika, logenomika, chemiomika en mikrobiomika. Opsomming: Die primêre benadering geneem in hierdie tesis is die gebruik van netwerke en meerveranderlike data-ontleding metodes om hoë-dimensie datastelle te ontleed. Verdermeer, verskeie van die metodes begin om die gemeenskaplike grond tussen netwerke en meerveranderlike data-ontleding te verken. Dit blyk om 'n logiese progressie te wees, aangesien beide netwerke en meerveranderlike data-ontleding gefokus is op matriks-gebaseerde data modellering en dus gewortel is in liniêre algebra.

Consolidated bioprocessing (CBP) provides a cost effective cellulose processing strategy, in which enzyme production, substrate hydrolysis, and fermentation of sugars to ethanol are all carried out in a single step by microorganisms. For industrial-scale bioethanol production, CBP-enabling microbes must be able to both efficiently degrade lignocellulosic material to fermentable sugars and synthesize bioethanol with high yields. Microbes with these properties have so far not been identified. Developing naturally occurring cellulolytic isolates with CBP-relevant properties requires a comprehensive understanding of their lignocellulosic hydrolysis mechanism and metabolism. In my quest to find a suitable organism for potential use in CBP, I took to investigate the under-characterized anaerobic bacterium, Clostridium termitidis strain CT1112. C. termitidis produces fermentative hydrogen and ethanol from a variety of lignocellulose derived substrates. I sought to investigate the metabolism of C. termitidis on different substrates and the mechanisms of substrate hydrolysis using a combination of microscopy, comparative bioinformatics, and ‘Omic (transcriptomic and proteomic) analyses. Comparative bioinformatics analyses revealed higher numbers of genes encoding carbohydrate active enzymes (CAZymes) with the potential to hydrolyze a wide-range of carbohydrates, and ‘Omic analyses were used to quantify the levels of expression of CAZymes, including endoglucanases, exoglucanases, hemicellulases and cellulosomal components. While cellulases and cellulosome components were highly expressed on cellulose, xylanases and glucosidases were predominantly expressed on pentoses, and chitinases (as well as cellobiose phosphorylases) were significantly up-regulated on cellobiose. In addition to growth on xylan, the simultaneous consumption of two important lignocellulose constituents, cellobiose and xylose was also observed. The ability to metabolize both hexose and pentose sugars is a highly desirable feature of CBP-relevant organisms. Metabolic profiles in association with ‘Omics analyses showed that hexoses and pentoses are consumed via the Embden-Meyerhof-Parnas and Pentose-Phosphate pathways, respectively, and that the genome content and expression profiles dictate end-product synthesis patterns. Genes and gene-products of enzymes in central metabolism and end-product synthesis were detected in high abundance under all substrate conditions, regardless of the amounts of end-products synthesized. The capabilities described thus far, identifies C. termitidis as a strain of interest for CBP. Further studies are, however, required for its development in to an industry-ready strain for biofuel production. / February 2016

Traditional water monitoring encounter limitations due to the large number of contaminants present in our waters possible giving raise to mixture effects. This thesis aimed to investigate how the emerging omics approaches (transcriptomics, proteomics and metabolomics) can be used as an effect-based monitoring approach to assess and predict adverse effects in the freshwater environment. Moreover, this thesis analysed challenges and barrier with omics. A systematic literature search was conducted using Scopus and Web of Science to find case-studies using omics in field studies and reviews regarding challenges and barriers. The results in this thesis suggest that the use of fish species (either collected in the wild or in situ set-ups), transcriptomics and investigations of WWTP recipient was the most common way to apply omics. In order to interpret omics-data multiple studies conducted chemical monitoring in conjunction, investigated additional traditional biomarkers and/or used omics to identify altered biological or functional pathways that possible could lead to adverse effects at higher levels. According to the challenges and barriers identified in this thesis, the future of omics in environmental monitoring rely on the possibility to characterise and quantify natural variability, define appropriate critical effect sizes (i.e. thresholds of critical effects) and define baseline data. Moreover, it is necessary to develop frameworks and standardisations for omics-approaches (e.g. study-designs) to promote the interpretation of the results. Future research is also needed to develop and increase the understanding of how the proteomics and metabolomics can be applied. By improving the use of omics a more holistic water monitoring can be achieved including screenings for biological responses and the ability to detect early warnings which will enhance the prioritisation and site management of polluted water bodies, including those with limited prior knowledge regarding potential contaminants.