Strains MS455 and MS389, endophytic bacteria, were isolated from healthy soybean plant growing adjacent to a patch of plants affected by charcoal rot disease, caused by the fungal pathogen Macrophomina phaseolina. The complete genomes of both strains were sequenced and identified as Burkholderia species Strain MS455 exhibits broad-spectrum antifungal activities against economically important pathogens, including Aspergillus flavus. Random and site-specific mutations were employed in discovery of the genes that share high homology to the ocf gene cluster of Burkholderia contaminans strain MS14, which is responsible for production of the antifungal compound occidiofungin. RNA-seq analysis demonstrated ORF1, a homolog to the ambR1 LuxR-type regulatory gene, not only regulates occidiofungin biosynthesis in MS455, but also involved in expression of multiple genes, especially those involved in ornibactin biosynthesis. Plate and corn kernel assays showed that growth of A. flavus and aflatoxin production were reduced significantly by MS455 as compared with buffer control and the ORF1 mutant. Strain MS389 shows significant antifungal and antibacterial activities as well. Mutagenesis study identified that the TatC gene, an important unit of twin-arginine translocase (Tat) secretion system, and the LysR-type transcriptional regulatory gene were essential for the antifungal activity of strain MS389. RNA-seq analysis implied that the pyrrolnitrin biosynthesis gene cluster and an uncharacterized NRPS / PKS gene cluster were involved in antifungal activity. By comparing several endophytic bacteria of Burkholderia, including MS455 and MS389, to pathogenic Burkholderia species, endophytic bacteria were observed to harbor multiple antimicrobial biosynthesis genes but lack certain pathogenic or virulence genes. The potential endophytic behavior related genes and characteristics related to antibiotic resistance, secretion system, and CRISPR-Cas profiles were well established. The research findings on strains MS455 and MS389 have provided important genetic clues for understanding their molecular mechanism of antimicrobial activities and exhibited their potential possibility as biocontrol agents.

Small bacterial genomes are believed to be evolutionarily derived from larger genomes through massive loss of genes and are usually associated with symbiotic or pathogenic lifestyles. It is therefore intriguing that a similar phenomenon of genome reduction has been reported within a group of free-living phototrophic marine cyanobacteria Prochlorococcus. Here I have investigated the roles of natural selection and mutation rate in the process of Prochlorococcus genome size reduction. Using a data set of complete cyanobacterial genomes including 12 Prochlorococcus and a sister group of 5 marine Synechococcus, I first reconstructed the steps leading to Prochlorococcus genome reduction in a phylogenetic context. The result reveals that small genome sizes within Prochlorococcus were largely determined by massive gene loss shortly after the split of Prochlorococcus and Synechococcus (a process we refer to as early genome reduction). A maximum likelihood approach was then used to estimate changes in both selection effect and mutation rate in the evolutionary history of Prochlorococcus. I also examined the effect of selection and functional importance of a subset of ancestor-derived genes those are lost in Prochlorococcus but are still retained in the genomes of its sister Synechococcus group. It appears that purifying selection was strongest when a large number of small effect genes were deleted from nearly all functional categories. And during this period, mutation rate also accelerated. Based on these results, I propose that shortly after Prochlorococcus diverged from its common ancestor with marine Synechococcus, its population size increased quickly and thus the efficacy of selection became very high. Due to limited nutrients and relatively constant environment, selection favored a streamlined genome for maximum economies in material and energy, causing subsequent reduction in genome size and possibly also contributing to the observed higher mutation rate.

Aufgrund von Fortschritten im Bereich der DNA-Sequenzierung hat die Anzahl verfügbarer Genome in den letzten Jahrzehnten rapide zugenommen. Tausende bereits heute zur Verfügung stehende Genome ermöglichen detaillierte vergleichende Analysen, welche für die Beantwortung relevanter Fragestellungen essentiell sind. Dies betrifft die Assoziation von Genotyp und Phänotyp, die Erforschung der Besonderheiten komplexer Proteine und die Weiterentwicklung medizinischer Anwendungen. Um all diese Fragen zu beantworten ist es notwendig, proteinkodierende Gene in neu sequenzierten Genomen zu annotieren und ihre Homologieverhältnisse zu bestimmen. Die bestehenden Methoden der Genomanalyse sind jedoch nicht für Menge heutzutage anfallender Datenmengen ausgelegt. Daher ist die zentrale Herausforderung in der vergleichenden Genomik nicht die Anzahl der verfügbaren Genome, sondern die Entwicklung neuer Methoden zur Datenanalyse im Hochdurchsatz. Um diese Probleme zu adressieren, schlage ich ein neues Paradigma der Annotation von Genomen und der Inferenz von Homologieverhältnissen vor, welches auf dem Alignment gesamter Genome basiert. Während die derzeit angewendeten Methoden zur Gen-Annotation und Bestimmung der Homologie ausschließlich auf codierenden Sequenzen beruhen, könnten durch die Einbeziehung des umgebenden neutral evolvierenden genomischen Kontextes bessere und vollständigere Annotationen vorgenommen werden. Die Verwendung von Genom-Alignments ermöglicht eine beliebige Skalierung der vorgeschlagenen Methodik auf Tausende Genome. In dieser Arbeit stelle ich TOGA (Tool to infer Orthologs from Genome Alignments) vor, eine bioinformatische Methode, welche dieses Konzept implementiert und Homologie- Klassifizierung und Gen-Annotation in einer einzelnen Pipeline kombiniert. TOGA verwendet Machine-Learning, um Orthologe von Paralogen basierend auf dem Alignment von intronischer und intergener Regionen zu unterscheiden. Die Ergebnisse des Benchmarkings zeigen, dass TOGA die herkömmlichen Ansätze innerhalb der Placentalia übertrifft. TOGA klassifiziert Homologieverhältnisse mit hoher Präzision und identifiziert zuverlässig inaktivierte Gene als solchet. Frühere Versionen von TOGA fanden in mehreren Studien Anwendung und wurden in zwei Publikationen verwendet. Außerdem wurde TOGA erfolgreich zur Annotation von 500 Säugetiergeenomen verwendet, dies ist der bisher umfangreichste solche Datensatz. Diese Ergebnisse zeigen, dass TOGA das Potenzial hat, sich zu einer etablierten Methode zur Gen-Annotation zu entwickeln und die derzeit angewandten Techniken zu ergänzen.

The human gut microbiome is home to many hundreds of different microbes which play a crucial role in human physiology. For most of them, little is known about how their genetic diversity translates into functional traits and how they interact with their host, which is to some extent due to the lack of isolate genomes. Cultivation-free metagenomic approaches yield extensive amounts of bacterial genetic data, and recently developed algorithms allow strain-level resolution and reconstruction of bacterial genomes from metagenomes, yet bacterial within-species diversity and transmission dynamics after fecal microbiota transplantation remain largely unexplored over cohorts and using these technological advances. To investigate bacterial within-species diversity I first undertook large-scale exploratory studies to characterize the population-level genomic makeup of the two key human gut microbes Eubacterium rectale and Akkermansia muciniphila , leveraging many hundreds of bacterial draft genomes reconstructed from short-read shotgun metagenomics datasets from all around the planet. For E. rectale , I extended previous observations about clustering of subspecies with geography, which suggested isolation by distance and the putative ancestral loss of four distinct motility operons, rendering a subspecies specifically found in Europe immotile. For A. muciniphila, I found that there are several closely related but undescribed Akkermansia spp. in the human gut that are all likely human-specific but are differentially associated with host body mass index, showcasing metabolic differences and distinct co-abundance patterns with putative cognate phages . For both species, I discovered distinct subspecies-level genetic variation in structural polysaccharide synthesis operons. Next, utilizing a complementary strain-resolved approach to track strains between individuals, I undertook a fecal microbiota transplantation (FMT) meta-analysis integrating 24 distinct clinical metagenomic datasets. I found that patients with an infectious disease or those who underwent antibiotic treatment displayed increased donor strain uptake and that some bacterial clades engraft more consistently than others. Furthermore, I developed a machine-learning framework that allows optimizing microbial parameters - such as bacterial richness - in the recipient after FMT based on donor microbiome features, representing first steps towards making a rational donor choice. Taken together, in my work I extended the strain-level understanding of human gut commensals and showcased that genomes from metagenomes can be suitable to conduct large-scale bacterial population genetics studies on other understudied human gut commensals. I further confirmed that strain-resolved metagenomics allows tracking of strains and thus inference of strain engraftment characteristics in an FMT meta-analysis, revealing important differences in engraftment over cohorts and species and paving the way towards better designed FMTs. I believe that my work is an important contribution to the field of microbiome research, showcasing the power of shotgun metagenomics, modern algorithms and large-scale data analysis to reveal previously unattainable insights about the human gut microbiome.

Clostridium perfringens is a significant human pathogen, causing 250,000 cases of food poisoning in addition to several thousand potentially lethal cases of gas gangrene each year in the United States. Historically, work in this field has centered around toxin production, as C. perfringens can produce over 13 toxins. This work expands the knowledge of the starvation-response of C. perfringens, which includes several potential virulence factors, sporulation, motility and biofilm formation. Sporulation protects cells from a variety of stresses, including starvation. Efficient sporulation requires the transcriptional regulator CcpA, mediator of catabolite repression. Sporulation is repressed by glucose, but, surprisingly, in a CcpA-independent fashion. C. perfringens cells in a biofilm are resistant to a number of environmental stresses, including oxygen and antibiotics. Biofilm formation is repressed by glucose, and other carbohydrates, independently of CcpA. Gliding motility, a type four pili (TFP)-dependent phenomenon, affords C. perfringens with a mechanism for moving across a solid surface in response to carbohydrate starvation, while carbohydrates supplementation at high levels delay the initiation of the motility response. CcpA is required for the proper initiation of motility, a ccpA^-C. perfringens strain showed a considerable increase in the time to initiation of motility on lactose and galactose, and was unable to move at all in the presence of glucose. Gliding motility represents the most significant finding of this work. TFP were previously undescribed in any Gram-positive bacterial species, and this work produced genetic evidence suggesting their presence in all members of the clostridia, and physical evidence for TFP-dependent gliding motility in a second species, C. beijerinckii. / Ph. D.

Fungi play important roles in various environments. Some of them infect plants and cause economically important diseases. However, many fungal pathogens cause similar symptoms or are even spread asymptomatically, making it difficult to identify them morphologically. Therefore, culture-independent, sequence-based diagnostic methods that can detect and identify fungi independently of the symptoms that they cause are desirable. Whole genome metagenomic sequencing has the potential to enable rapid diagnosis of plant diseases without culturing pathogens and designing pathogen-specific probes. In my study, the MinION nanopore sequencer, a portable single‐molecule sequencing platform developed by Oxford Nanopore Technologies, was employed to detect the fungus Calonectria pseudonaviculata (Cps), the causal agent of the devastating boxwood blight disease of the popular ornamental boxwood (Buxus spp.). Various DNA extraction methods and computational tools were compared. Detection was sensitive with an extremely low false positive rate for most methods. Therefore, metagenomic sequencing is a promising technology that could be implemented in routine diagnostics of fungal diseases. Other fungi may play important roles in the atmosphere because of their ice nucleation activity (INA). INA is the capacity of some particles to induce ice formation above the temperature that pure water freezes (-38°C). Importantly, INPs affect the ratio of ice crystals to liquid droplets in clouds, which in turn affects Earth's radiation balance and the intensity and frequency of precipitation. A few fungal species can produce ice nucleating particles (INPs) that cause ice formation at temperatures ≥ –10°C and they may be present in clouds. Two such fungal genera are Fusarium and Mortierella but little is known about their INPs and the genetic basis of their INA. In my study, F. avenaceum and M. alpina were examined in detail. INPs of both species were characterized and it was found that strains within both species varied in regards to the strength of INA. Whole genome sequencing and comparative genomic studies were then performed to identify putative INA genes. Differential expression analyses at different growth temperatures were also performed. INP properties of the two species shared similarities, both appearing to consist of secreted aggregates larger than 30 kDa. Low temperatures induced INA in both species. Lists of candidate INA genes were identified based on their presence in the strains with the strongest INA and/or induction of their expression at low temperatures and because they either encode secreted proteins or enzymes that produce other molecules known to have INA in other organisms. These genes can now be characterized further to help identify the fungal INA genes in both species. This can be expected to help increase our understanding of the role of fungal INA in the atmosphere. / Doctor of Philosophy / Fungi are important to life on Earth and play roles in the environments that surround us. On the one hand, fungi can make plants sick and some plant diseases may even cause economic losses to farmers. If the cause of a disease can be identified accurately in an early stage before symptoms develop, disease transmission may be prevented and plants may be protected from disease. However, it is a challenge to find out which fungus causes which disease since symptoms of different fungal diseases look very similar. Typically, we have to wait for plants to become very sick or we have to isolate the fungus that causes a disease to identity it, which may be time-consuming and not lead to precise identification. DNA sequencing technologies have the potential to lead to more sensitive, faster, and more accurate disease diagnosis and, therefore, may help prevent disease outbreaks. In my study, the MinION nanopore sequencer, a small portable device, was used to detect the fungus causing boxwood blight on boxwood. By loading the DNA of unhealthy boxwood on the device, the boxwood blight pathogen was identified within a very short time. Thus, this method is a promising diagnostic method that may be applied to detect other plant fungal diseases as well. On the other hand, fungi may affect Earth's climate by affecting how many water droplets in clouds are frozen, which in turn affects Earth's temperature and how often and how much it rains and snows. Fungi may affect the freezing of water droplets in clouds since some of them have ice nucleation activity (INA), which is the capacity to catalyze ice formation at a higher temperature than the temperature at which pure water freezes (-38°C), and they may be present in clouds. So far, INA has only been found in a few fungi, including the species Fusarium avenaceum and Mortierella alpina, but the mechanism of their INA is poorly understood. In my study, multiple F. avenaceum and M. alpina strains were examined in detail. Two approaches were used. First, strains in each species were compared with each other to find out how strong their INA is. Once it was found that they differed in their strength of INA, their genomes were sequenced and compared to find genes present in the most active strains and missing from the least active strains since it is these genes that may contribute to INA. It was also found that both fungal species had stronger INA when they were grown at lower temperatures. Therefore, the expression of their genes between higher and lower temperatures was compared to find the genes that were more highly expressed at lower temperatures since it is these genes that may cause INA. Based on previous studies, fungal INPs may either consist of secreted proteins or be the products of biosynthetic gene clusters. Therefore, the list of potential genes was reduced by looking for genes encoding either secreted proteins or biosynthetic gene clusters. The list of these potential INA genes will make it easier to identify the INA genes in F. avenaceum and M. alpina and determine the role of fungi in affecting the weather and climate on Earth.

The variation of genome content and structure across the tree of life is astounding and can provide clues to understand the process of evolution. Overall, this helps us understand the history of life and how organisms have fundamentally changed and adapted to their environments. Gene duplication is an important mechanism for molecular evolution because it provides opportunity for functional novelty and molecular innovation. Gene duplication creates new functional gene copies with different selective pressures that allow them to take on new or specialized functions. Throughout this work, I explored the interplay between genetic changes, molecular phenotype, and the selection of duplicate gene copies. I particularly focused on the genetic opportunity, consequences, and selective pressures of the mechanisms for short-term and long-term duplicate copy retention. I modeled the stochastic processes of mutation and selection and their effect on duplicate gene copy retention. Specifically, I modeled the interplay between subfunctionalization and dosage balance and found that selection may cause genes that are sensitive to dosage balance effects to experience delayed subfunctionalization, but ultimately lead to higher levels of subfunctionalization. These findings suggest that subfunctionalization may not occur as a purely neutral process. Next, I used survival analysis methods to model patterns of duplicate gene retention in genomes experiencing consecutive whole genome duplication events. I modeled three hypotheses to explain patterns of duplicate gene retention including the Independence Hypothesis, the Gene Duplicability Hypothesis, and a novel Mutational Opportunity Hypothesis. Under the Gene Duplicability and Mutational Opportunity hypotheses, the expected patterns of duplicate gene retention after consecutive whole genome duplication events are greatly affected by the ages of the whole genome duplication events and the functional properties of the genomic content that influence opportunity and selection. Additionally, I describe how statistical model testing techniques can be applied to investigate which hypothesis is consistent with patterns of retention in real-world phylogenetic datasets. I used these described techniques to explore the hypotheses’ parameter space consistent with a modest dataset of fish and plant lineages. These results suggest that a gene duplicate’s retention after whole genome duplication events may be influenced by its functional properties. Key findings underscore the multifaceted nature of duplicate gene retention, influenced by a myriad of factors including genetic opportunity, selective pressures, and evolutionary context. By dissecting the underlying mechanisms driving duplicate gene retention, this dissertation advances our understanding of the evolutionary dynamics shaping genome evolution and functional diversity across diverse biological systems. / Biology

In eukaryotes, recombination plays a critical role in both the production of viable gametes and as a population genetic process. Here, we are interested in studying recombination as it provides insight into a process that has shaped variation. To this end, we study the evolution of cross-over rates in chimpanzees and humans through two experiments. Components of the recombination machinery are well described in yeast and C. elegans, but less so in other species. In humans, cross-over rates vary across physical scales and occur predominantly in narrow ∼2 kb regions called hotspots, where hotspot usage differs considerably between individuals. Differential hotspot usage is associated with specific DNA motifs, and DNA-contacting zinc finger array variants in the transacting PRDM9 H3K4 trimethyltransferase. The precise relationship between DNA motifs, PRDM9 and hotspot activity is not completely understood. Experiment 1. To investigate the importance of PRDM9 motif recognition, which is predicted be different between humans and chimpanzees, and the effect of PRDM9 on the evolution of fine-scale cross-over rates, we sequenced 10 unrelated Pan troglodytes verus (Western chimpanzee) genomes to moderate coverage (∼10×). I validate the approach by demonstrating that fine-scale maps estimated from 10 human genomes of each African and European ancestry recapitulate independently estimated maps. Then I characterise the error modes in sequencing data arising from errors in chemistry, alignment, variant calling, and genotyping. I identify several cryptic error modes missed by state-of-the-art filters and develop methods to counteract them. To guard against genotype error arising from stochastic variation in low to moderate coverage sequencing, I develop methods to incorporate the underlying statistical uncertainty into recombination analyses, evaluate the approaches through simulation (estimated 11% improvement) and empirical assessment (estimated 4% improvement), and discover that the reported genotype uncertainty is poorly calibrated, which limits the approaches. Consequently, a filtering approach was applied to the hard-called chimpanzee genotypes. I estimate recombination rates in chimpanzees through an existing LD-based method. In contrast to humans, there is no increased cross-over localisation around chimpanzee PRDM9 binding predictions, nor motifs consistently associated with activity. Hotspots do not overlap between the two species, indicating that rates evolved rapidly and consistent with PRDM9 localising all hotspots. In contrast, gene pro- moters and CpG islands are common attractors of recombination (2.7-fold increase in rate in chimpanzee, 1.5-fold increase in human), suggesting chromatin state influences hotspot placement but to varying degree in the species. I discuss the potential implications for PRDM9 mechanism. Experiment 2. To enable a more representative characterisation of the spectrum of genome changes occurring in chimpanzee genomes, I analyse data from an extended three generation Western chimpanzee pedigree sequenced at high coverage (∼30×). I use Mendel transmission to filter variants, infer haplotypes, and identify recombination events through a Hidden Markov Model approach. We detect 375 recombination events, of which 3 are double cross-over events. Sex-specific recombination rate estimates in chimpanzees mirror sex differences in humans (N♂/N♀ = 0.58) and have similar levels of total recombination. We resolve recombination events typically at ∼ 856 base-pair resolution. Additionally, analyses of Mendel inconsistencies suggest that extended pedigree sequencing opens the door on studying complex genome changes. These experiments demonstrate the power of comparative analyses, the utility of high throughput sequencing in enabling the study of recombination in almost any species of interest, the challenges in sifting signal from noise in these data, and the need for experimental and algorithmic methods to guard against error.

Endocytosis is both an ancient and a diverse feature of the eukaryotic cell. Studying how it evolved can provide insight into the nature of the last common eukaryotic ancestor, and the diversification of eukaryotes into the known extant lineages. In this thesis, I present two studies on the evolution of endocytosis. In the first part of the thesis I report results from a large-scale, phylogenetic and comparative genomic study of clathrin-mediated endocytosis (CME). The CME pathway has been studied to a great level of detail in yeast to mammal model organisms. Several protein families have now been identified as part of the complex set of protein-protein and protein-lipid interactions which mediate endocytosis. To investigate how such complexity evolved, first, I defined the modular nature of the CME interactome (CME-I) by literature review, and then I carried out a systematic phylogenetic and protein domain architecture analysis of the proteins involved. These data were used to construct a model of the evolution of the CME-I network, and to map the expansion of the network's complexity to the eukaryotic tree of life. In the second part of the thesis, I present results from evolutionary and functional studies of the eisosome, a protein complex which has been proposed to regulate the spatial distribution of endocytosis in S. cerevisiae. The phylogeny of eisosomes components Pil1 and Lsp1 reported here, suggests that eisosomes are likely to have originated at the base of the fungi, and then diversified significantly via multiple gene duplications. I thus studied the localisation and function of Pil1 and Lsp1 homologues in Magnaporthe oryzae to investigate the role of eisosomes in filamentous fungi. Results suggests that eisosomes are linked with septal formation and integrity in M. oryzae, and that the septal specific Pil2 paralogue was lost in budding yeasts. Together, the data presented in this thesis describe the evolutionary history of a complex biological system, but also highlights the problem of asymmetry in the understanding of endocytic diversity in the eukaryotes.

Pour mieux saisir les liens complexes entre génotype et phénotype, une méthode utilisée consiste à étudier les relations entre différents éléments biologiques (entre les protéines, entre les métabolites...). Celles-ci forment ce qui est appelé un réseau biologique, que l'on représente algorithmiquement par un graphe. Nous nous intéressons principalement dans cette thèse au problème de la recherche d'un motif (multi-ensemble de couleurs) dans un graphe coloré, représentant un réseau biologique. De tels motifs correspondent généralement à un ensemble d'éléments conservés au cours de l'évolution et participant à une même fonction biologique. Nous continuons l'étude algorithmique de ce problème et de ses variantes (qui admettent plus de souplesse biologique), en distinguant les instances difficiles algorithmiquement et en étudiant différentes possibilités pour contourner cette difficulté (complexité paramétrée, réduction d'instance, approximation...). Nous proposons également un greffon intégré au logiciel Cytoscape pour résoudre efficacement ce problème, que nous testons sur des données réelles.Nous nous intéressons également à différents problèmes de génomique comparative. La démarche scientifique adoptée reste la même: depuis une formalisation d'un problème biologique, déterminer ses instances difficiles algorithmiquement et proposer des solutions pour contourner cette difficulté (ou prouver que de telles solutions sont impossibles à trouver sous des hypothèses fortes) / To investigate the complex links between genotype and phenotype, one can study the relations between different biological entities. It forms a biological network, represented by a graph. In this thesis, we are interested in the occurrence of a motif (a multi-set of colors) in a vertex-colored graph, representing a biological network. Such motifs usually correspond to a set of elements realizing a same function, and which may have been evolutionarily preserved. We follow the algorithmic study of this problem, by establishing hard instances and studying possibilities to cope with the hardness (parameterized complexity, preprocessing, approximation...). We also develop a plugin for Cytoscape, in order to solve efficiently this problem and to test it on real data.We are also interested in different problems related to comparative genomics. The scientific method is the same: studying problems arising from biology, specifying the hard instances and giving solutions to cope with the hardness (or proving such solutions are unlikely)