My dissertation addresses two aspects of eukaryotic evolution, 1) the organization of eukaryotic diversity and 2) genomic variation in Foraminifera. The bulk of eukaryotic diversity is microbial with plants and animals representing just two of the estimated 75 lineages of eukaryotes. Among these microbial lineages, there are many examples of dynamic genome processes. Elucidating the origin and evolution of genome features requires a robust phylogenetic framework for eukaryotes. Taxon-rich molecular analyses provide a mechanism to test hypothesized evolutionary relationships and enable placement of diverse taxa on the tree of life. These analyses result in a well-resolved eukaryotic tree of life. Relaxed molecular clock analyses of this taxon-rich dataset place the origin on eukaryotes in the Paleoproterozoic, and suggest that all of the major lineages of eukaryotes diverged before the Neoproterozoic. This robust scaffold of the tree of eukaryotes is also used to elucidate common themes in genome evolution across eukaryotes. Mapping dynamic genome features onto this tree demonstrates that they are widespread in eukaryotes, and suggests that a common mechanism underlies genome plasticity. Foraminifera, a diverse lineage of marine amoebae, provide a good model system for investigating genome dynamics because they amplify portions of their genome and go through ploidy cycles during their life cycle. Assessment of nuclear dynamics in one species of Foraminifera, Allogromia laticollaris strain CSH, reveals that genome content varies according the life cycle stage and food source, which may differentially impact organismal fitness. The inclusion of diverse microbial eukaryotes enables better resolution of eukaryotic relationships and improves our understanding the dynamic nature of eukaryotic genomes.

tRNA genes are highly transcribed and perform one of the most fundamental cellular functions. Although a universal pattern observed across all three domains of life is that highly transcribed genes tend to evolve slowly, tRNA genes have been shown previously to evolve rapidly. This rapid sequence evolution could result from relaxed selection, increased mutation rate, or a combination of both. Here, we use mutation-accumulation line sequencing data to show that tRNA genes accumulate more mutations than other gene types. Our results indicate that this elevated mutation rate is a consequence of both elevated transcription-associated mutagenesis and a lack of transcription-coupled repair in tRNA genes. We also identify the gene MSH2 as being involved in transcription-coupled repair.

<p>Repetitive elements, including tandem repeats and transposable elements (TE), are genetic features of all plant and animal genomes. Despite their abundance and the phylogenetic breadth of host genomes, factors that control the genome-wide distribution of repetitive elements are not well understood. Here we have evaluated the correlation between various genomic predictor variables such as gene expression level, distance from genes, and GC content, with the presence of TEs and non-TE repeats in two kilobase windows of the complete genome sequence of the Western Clawed Frog (<em>Silurana tropicalis</em>). We found that the distributions of different classes of TEs and repeats have distinct correlations with these predictor variables, including a generally strong negative correlation with proximity to exons and GC content. We also found that DNA transposons, but not retrotransposons, are preferentially inserted or preferentially retained near germline-expressed genes. Retrotransposons and simple repeats are found more often in or near conserved regions than expected by chance. These results offer insights into various models that have been proposed to account for heterogeneity in the genomic distribution of repetitive elements, most notably for the “gene disruption model” which posits that TE insertion and repeat presence near or in genes imposes costs to host fitness. In general, multiple lines of evidence suggests that the nature of natural selection on TE and other repetitive element evolution in this frog appears to be similar to that acting on TE and other repetitive elements in the human genome. This is possibly related to the similar size and level of complexity of the genomes of both of these species.</p> / Master of Science (MSc)

African clawed frogs (Xenopus) have been scientific and medical model species for decades. These frogs present many curious features, and their genomic history is no exception. As such, a variety of evolutionary genomic questions can be addressed with these species in a comparative framework, owing to the great array of genetic tools available and a large number of abundant species. The sex chromosomes of this group are evolutionarily young, and this thesis establishes that there has been an additional change in what constitutes the sex chromosomes in one species of Xenopus. This allows us to compare the evolutionary trajectory of newly established sex chromosomes. By exploring the genetic content of these systems, profiling their recombinational activity, and assessing the extent of nucleotide divergence between the sex chromosomes, we find that sex chromosome evolution may be predictable in some aspects, and highly unpredictable in others. In addition, this genus is uncharacteristic for vertebrates in the frequency with which lineages undergo whole genome duplication. In this thesis, we explore the selective dynamics operating on duplicate genes over time, and the rate at which duplicate copies are purged from the genome from multiple Xenopus species. These investigations provide an animal perspective on the subject of biased subgenome evolution, characteristic of allopolyploids. The last two chapters of this thesis redefine the species boundaries for the most intensively studied Xenopus species (X. laevis), and explore the genetic extent of hybridization between the common X. laevis and the endangered X. gilli. Overall, this thesis provides a broad look at several aspects of Xenopus evolutionary genomics, providing novel contributions to the fields of sex chromosome research, whole genome duplication, and speciation and hybridization. / Thesis / Doctor of Philosophy (PhD)

The ocean plays essential roles in Earth's biochemistry. Most of the nutrient transformations that fuel trophic webs in the ocean are mediated by microorganisms. The extent of phylogenetic and metabolic diversity of key culture and uncultured marine microbial clades started to be revealed due to progress in sequencing technologies, however we still lack a comprehensive understanding of the evolutionary processes that led to the microbial diversity we see in the ocean today. In this dissertation, I apply phylogenomic and comparative genomic methods to explore the evolutionary genomics of bacterial and archaeal clades that are relevant due to their abundance and biogeochemical activities in the ocean. In Chapter 1, I review relevant literature regarding the evolutionary genomics of marine bacteria and archaea, with emphasis on the origins of marine microbial diversity and the evolution of genome architecture. In Chapter 2, I use a comparative framework to get insights into the evolutionary forces driving genome streamlining in the Ca. Marinimicrobia, a clade widely distributed in the ocean. This project shows that differences in the environmental conditions found along the water column led to contrasting mechanisms of evolution and ultimately genome architectures. In Chapter 3, I assess the phylogenetic signal and congruence of marker genes commonly used for phylogenetic studies of bacteria and archaea and propose a pipeline and a set of genes that provide a robust phylogenetic signal for the reconstruction of multi-domain phylogenies. In Chapter 4, I apply a phylogeny-based statistical approach to evaluate how tightly genome size in bacteria and archaea is linked to evolutionary ii history, including marine clades. I present evidence suggesting that phylogenetic history and environmental complexity are strong drivers of genome size in prokaryotes. Lastly, in Chapter 5, I estimate the emergence time of marine bacterial and archaeal clades in the context of the Prokaryotic Tree of Life and demonstrate that the diversification of these groups is linked to the three main oxygenation periods occurring throughout Earth's history. I also identify the metabolic novelties that likely led to the colonization of marine realms. Here I present methodological frameworks in the fields of comparative genomics and phylogenomics to study the evolution of marine microbial diversity and show evidence suggesting that the main evolutionary processes leading to the extant diversity seen in the ocean today are intimately linked to geological and biological innovations occurring throughout Earth's history. / Doctor of Philosophy / The ocean plays essential roles in the functioning of our planet. Many of the nutrient's transformation happening in marine environments are mediated by microorganisms, whose metabolic activities underpin higher trophic levels. The identity of the most prevalent marine microbial groups has been reveled during the last two decades through sequencing technologies. Despite having a great progress in our understanding of the functions that these microorganisms have in the ocean; we still lack information about the evolutionary processes that allowed their diversification and colonization into marine realms. In this work, I developed and applied computational strategies to disentangle the evolutionary genomics of marine microorganisms. One particularity about most these marine groups is that they have very small genomes. To explore the evolutionary forces driving their genome reduction, I analyzed a broad set of genomes of Marinimicrobia, a bacterial group widely distributed in the ocean. This analysis shows that genome reduction in Marinimicrobia is driven by negative selection, an evolutionary force that allows the deletion of non-essential genes, subsequently leading to genome reduction. Moreover, I developed a benchmarked pipeline for the reconstruction of phylogenetic trees to study the evolutionary relationships of microorganisms. This pipeline allowed me to link the diversification of the main marine groups and the geological periods in which they first emerged. I discovered that the colonization of these groups happened during three different periods, which are coincident with the main oxygenation events occurring across Earth's history. Moreover, the diversification of vi marine microbial groups was associated with the acquisition of genes to exploit the newly created niches that followed the oxygenation of the atmosphere and the ocean. Overall, my work shows that the diversification of the marine microbial clades that are essential for the functioning of the ocean is intimately linked to the redox state of the ocean and the atmosphere throughout Earth's history.

Les génomes des eucaryotes et des procaryotes évoluent de temps en temps par un processus complexe, impliquant entre autres, des événements évolutifs tels que les spéciations, les duplications, les transferts horizontaux, et les pertes de gènes. Nous étudions ici les méthodes de réconciliation, une technique bien connue pour inférer de tels événements et retrouver leur localisation dans l'histoire d'espèces. En effet, ces méthodes construisent une correspondance entre l'histoire d'une famille de gènes (l'arbre de gènes) et l'histoire des espèces contenant ces gènes (l'arbre d'espèces) pour expliquer leurs discordances sur la base d'événements évolutifs qu'elles infèrent et positionnent sur l'arbre de gènes et l'arbre d'espèces. Les méthodes de réconciliation sont appliquées dans plusieurs domaines tels que l'étude de l'évolution du génome; l'inférence des relations d'orthologies en évolution moléculaire; l'étude de la coévolution entre hôtes et parasites en écologie, ou encore l'étude des zones de population en biogéographie. Les trois principales contributions de cette thèse sont les suivantes : premièrement, un outil nommé SEAS est proposé pour simuler l'évolution des familles de gènes dans une phylogénie d'espèces donnée. Cela permet d'obtenir des arbres de gènes synthétiques dont la réconciliation est connue et qui permettent donc d'évaluer la précision des méthodes de réconciliation. Deuxièmement, une méthode heuristique, appelée MowgliNNI, est proposée pour corriger les arbres de gènes partiellement erronés au regard des réconciliations. Cette méthode itérative réarrange les branches faiblement supportées pour rechercher une nouvelle topologie de l'arbre de gènes, dont le coût de réconciliation est moindre. Troisièmement, nous proposons une approche pour estimer la fiabilité des événements évolutifs prédits par les méthodes de réconciliation. Contrairement aux approches existantes qui ne considèrent qu'une des réconciliations optimales possible entre l'arbre de gènes et l'arbre d'espèces, notre approche prend en compte un ensemble de solutions optimales voire sous-optimales. En outre, nous introduisons le concept de réconciliations médianes symétriques et asymétriques qui servent d'éléments centraux pour représenter un ensemble de réconciliations. Nous présentons un algorithme pour calculer ces réconciliations médianes qui est en temps polynomial bien que l'ensemble de toutes les réconciliations optimales est potentiellement exponentiel. Des expériences ont été réalisées pour montrer l'exactitude, la signification et l'efficacité de nos méthodes proposées. / The genomes of eukaryotes and prokaryotes evolve over time through a complex process involving, among other things, evolutionary events such as speciations, duplications, horizontal transfers, and losses of genes. We study here reconciliation methods, a well-known technique for recovering such events as well as locating them along the species history. Indeed, reconciliation methods construct a mapping between a gene family history (a gene tree) and a species history (a species tree) to explain their incongruence thanks to the inferred evolutionary events located on both the gene and species trees. Reconciliation methods can be applied to various areas such as the study of genome evolution, the inference of orthology relationships in molecular evolution, the study of host-parasite coevolution in ecology, or the study of population areas in biogeography. The three main contributions of this thesis are as follows: First, we provide a tool, named SEAS, for simulating the evolution of gene families along a given species phylogeny. This provides synthetic gene trees along with their known reconciliations that are helpful to evaluate the accuracy of reconciliation methods. Second, we propose a heuristic method, called MowgliNNI, to correct partly erroneous gene trees based on reconciliation scores. This method iteratively rearranges the weakly supported parts of a gene tree as long as it improves the reconciliation score. Third, we propose effective solutions for estimating the reliability of the predicted evolutionary events. Unlike the currently existing approaches considering only the optimal solutions for reconciling a pair of species-gene trees, our approach additionally takes into account the nearly optimal solutions. Furthermore, we introduce the concept of symmetric and asymmetric median reconciliations, which serve as central elements to represent a set of reconciliations. We present a polynomial time algorithm computing such median reconciliations from the potentially exponential set of all optimal reconciliations for a given pair of species-gene trees. Experiments have been carried on to show the correctness, meaningfulness and effectiveness of our proposed methods.

Os elementos de transposição (TEs) se movimentam de um locus para outro no genoma afetando a estrutura e evolução destes. A superfamília de transposases hAT é definida pelos elementos que compartilham os domínios de dimerização e ligação ao DNA com os transposons previamente descritos: hobo, Activator e Tam3. Análises prévias encontraram algumas evidencias da presença genômica e ativação transcricional de TEs relacionados à superfamília hAT em cana-de-açúcar (denominados de família SChAT) e pelo menos três linhagens evolutivas foram postuladas. O objetivo deste trabalho é caracterizar versões genômicas das linhagens de transposons (191 e 257) e linhagem possivelmente domesticada (074). Pretende-se estudar as relações evolutivas, distribuição em gramíneas, identificar os padrões de expressão e propriedades funcionais. Regiões de sintenia foram estabelecidas para estes BACs em Arabidopsis thaliana, Brachypodium distachyon, Sorghum bicolor, Oryza sativa e Zea mays. Elementos relacionados com as três linhagens foram procurados nestes genomas. / Transposable elements (TEs) are able to move from one locus to another within a genome. TE mobilization affects genome structure and evolution. The hAT transposase superfamily is defined as elements that share the dimerization and DNA ligation domains with the previously described hobo, Activator and Tam3 transposon elements. Previous analyses found some genomic and transcriptional evidences of TEs related to hAT superfamily in sugarcane (named SChAT family) and at least three evolutionary lineages were proposed. The aim of this work is to characterize full-length genomic versions of the transposons lineages (191 and 257) and from the domesticated lineage (074). It is proposed to study the evolutionary relationship, distribution along grasses genomes, identify expression patterns and functional capacities of the SChAT elements. Syntenic regions for the BACs containing elements from the three lineages were mapped in Arabidopsis thaliana, Brachypodium distachyon, Sorghum bicolor, Oryza sativa and Zea mays. Related elements were search on the same genomes.

One of the most unforgiving processes in nature is that of Muller's ratchet, a seemingly irreversible accumulation of deleterious mutations that all organisms have to deal with or face extinction. The most obvious way to avoid fitness collapse is recombination, though asexual populations usually do not have the luxury of recombining freely.  With the aid of computational and mathematical models, we have studied other situations where this threat is averted and the organism can survive the ratchet. The results show that a ratchet where all mutations have the same deleterious fitness effect is very effectively stalled for large effects. However, if mutations are allowed to have a broad range of effects, the fitness-loss rate can be substantial even with the same mean effect as the one-type ratchet, but we have  identified parameter regions where even the broad-range effects are effectively stopped. The fitness-loss from a ratchet is very sensitive to the mutation rate and a mutation that increases the mutation rate (mutator) can easily start an otherwise stalled ratchet. Large effect mutators are heavily counter-selected, but smaller mutators can spread in the population. They can be stopped by reversals (antimutators), but even if the mutation rate is equilibrated in this way, there will be large fluctuations in mutation rate and even larger in the fitness-loss rate due to the feedback amplification in their coupling.    Another way of preventing the ratchet is by reversal of the deleterious mutations themselves through back-mutations or compensatory mutations. The rate required to stop the ratchet using only back-mutations before the fitness collapses is very large. A detailed comparison between the deleterious mutations in the ratchet and in a sexual population was made and the difference was found to be greatest for large populations with large genomes. There are obviously many ways to survive the ratchet, but even more ways to drive a species to extinction by enhancing and speeding up the ratchet. By modelling and testing the ratchet for numerous different situations, we show the effects of some of these threats and benefits.

A genome is shaped by many different forces. Recombination can for instance both create and maintain genetic diversity, but the need to locally reduce recombination rates will also leave specific signatures. Genetic elements can act selfishly and spreading at the expense of the rest of the genome can leave marks of their activity, as can mechanisms that suppresses them, in a phenomenon known as genomic conflict. In this thesis, I have studied the forces driving genome evolution, using modern genome sequencing techniques and with a special focus on a class of selfish genetic elements known as Spore killers found in the fungus Neurospora. First, we show novel findings on large-scale suppression of recombination by non-structural means in the N. tetrasperma genomes. In contrary, in the genomic region harbouring the spore killer elements Sk-2 and Sk-3 of N. intermedia, a dense set of inversions that are interspersed with transposable elements have accumulated. The inversions are unique for each killer type, showing that they have a long separated evolutionary history and likely have established themselves independently. For the Sk-2 haplotype, where we have polymorphism data, we see signs of relaxed selection, which is consistent with the hypothesis that recombination suppression reduces the efficacy of selection in this region. These results show the strong effects the divergent selective forces of genomic conflicts can have on chromosome architecture. Furthermore, we investigate the hypothesis that spore killing can drive reproductive isolation, by comparing the fertility of crosses between N. metzenbergii and either killer or non-killer N. intermedia strains. We show that crosses with spore killer strains have lower fertility, which cannot be explained by the killing itself, but is potentially caused by an incompatibility gene captured in the non-recombining region. Finally, we identified the genetic element responsible for causing spore killing in the Sk-1 spore killer strains found in N. sitophila. Unlike the Sk-2 and Sk-3 elements, Sk-1 is not connected to a large, non-recombining region, but is caused by a single locus, and we also find indications that this locus was introgressed from N. hispaniola.

La recombinaison méiotique via les Crossing-Overs (COs) est le principal mécanisme permettant le brassage de la diversité génétique. Cependant, le nombre et la position des COs entre paires de chromosomes homologues sont strictement régulés, limitant la séparation des loci en sélection variétale. Dans le cas du colza B. napus, l’utilisation d’allotriploïdes (AAC, 2n=3x=29), issus du croisement entre le colza (AACC, 2n=4x=38) et l’un de ses progéniteurs B. rapa (AA, 2n=2x=20), permet d’augmenter considérablement le nombre de COs entre chromosomes homologues A. L’objectif de cette étude était de déterminer les conséquences d’une telle variation sur la distribution des COs le long des chromosomes ainsi que d’identifier des facteurs régulant ce phénomène. Suite à la production et à la caractérisation cytogénétiques d’hybrides F1 présentant différents caryotypes, la recombinaison homologue a été évaluée par des analyses génétiques via des marqueurs SNPs physiquement ancrés sur l’ensemble duNous avons montré que l’addition du génome C chez les allotriploïdes conduit toujours à (1) la formation de COs surnuméraires, dont le nombre varie fonction des méioses mâle/femelle et du fond génétique, (2) une modification des profils de recombinaison, notamment au voisinage des centromères, et (3) une réduction de l’intensité d’interférence. De plus, nous avons révélé que le contrôle génétique de ces variations est imputé à des chromosomes C spécifiques et aurait divergé dans un contexte polyploïde. Nous avons donc identifié un levier permettant d’optimiser le brassage de la diversité gén / Meiotic recombination via crossovers (COs) is the main mechanism responsible for mixing genetic diversity. However, the number and position of COs between the pairs of homologous chromosomes are strictly regulated, limiting the loci separation in plant breeding. In the case of the rapeseed B. napus, the use of allotriploids (AAC, 2n=3x=29), resulting from the cross between rapeseed (AACC, 2n=4x=38) and one of its progenitors B. rapa (AA, 2n=2x=20), allows a substantial increase of the number of COs between homologous A chromosomes. The objective of this study was to determine the consequences of such a variation on the distribution of COs along the chromosomes and to identify factors regulating this phenomenon. Following the production and cytogenetic characterization of F1 hybrids with different karyotypes, homologous recombination was assessed by genetic analyzes via SNPs markers physically anchored on the whole A genome.We showed that the additional C genome in allotriploids always leads to (1) the formation of extra COs, for which the number depends on the male/female meiosis and the genetic background, (2) the modification of the recombination landscapes, especially in the vicinity of centromeres, and (3) the decrease of CO interference. In addition, we revealed that the genetic control of these variations is assigned to specific C chromosomes and could have evolved in a polyploid context. We have therefore identified a way to optimize the shuffling of genetic diversity in rapeseed breeding.