Loblolly pine (Pinus taeda L.) is an ecologically and economically important southern pine, distributed across the southeastern United States. Its genetic improvement for breeding and deployment is a major goal of the Western Gulf Forest Tree Improvement Program (WGFTIP) hosted by the Texas A&M Forest Service. Rapid advances in genomics and molecular marker technology have created potential for application of Marker Assisted Selection (MAS) and Genomic Selection (GS) for accelerated breeding in forest trees. First-generation selection (FGS) and second- generation selection (SGS) breeding populations of loblolly pine from east Texas were studied to estimate the genetic diversity, population structure, linkage disequilibrium (LD), signatures of selection and association of breeding traits with genetic markers using a genome-wide panel of 4264 single nucleotide polymorphisms (SNPs). Under- standing the genetic basis of local adaptation is crucial to disentangle the dynamics of gene flow, drift and selection and to address climate change. Bayesian mixed linear models and logistic regression were used to associate SNP variation with geography, climate, aridity and growth season length and markers with strong correlations were investigated for biological functions. Relatively high levels of observed (Ho = 0.178–0.198) and expected (He = 0.180-0.198) heterozygosities were found in all populations. The amount of inbreeding was very low, and many populations exhibited a slight excess of heterozygotes. The population substructure was weak, but FST indicated more pronounced differentiation in the SGS populations. As expected for outcrossing natural populations, the genome-wide LD was low, but marker density was insufficient to deduce the decay rate. Numerous associations were found between various phenotypes and SNPs, but few remained significant after false positive correction. Signatures of diversifying and balancing selection were found in markers representing important biological functions. Strong correlations supported by Bayes factors were found between various environmental variables and several SNPs. Logistic regression found hundreds of significant marker-environment associations, but none remained significant after false-positive correction, which was likely too stringent and will require further investigation. Annotations of significant markers implicated them in crucial biological functions. These results present the first step in the application of MAS to the WGFTIP for loblolly pine genetic improvement and will contribute to the knowledgebase necessary for genomic selection technology. Results from environmental association study provide important information for designing breeding strategies to address climate change and for genetic conservation purposes.

The American chestnut tree (Castanea dentata) is a deciduous tree that largely exists in the eastern United States along the Appalachian Mountain range. Approximately 100 years ago, a fungal pathogen (Cryphonectria parasitica) decimated chestnut populations, resulting in the loss of billions of trees. Disease-resistant American chestnut populations have been developed, but the introgression of wild adaptive diversity into these breeding populations will be necessary to develop locally adapted and disease resistant chestnut trees for reintroduction. In this dissertation, I presented our findings which addressed previous gaps in knowledge regarding the population genomics of wild and backcross American chestnut populations. I 1) estimated the genomic diversity, population structure, and demographic history of remnant wild American chestnut populations; 2) revealed the genomic basis of local climate adaptation in American chestnut, developed a novel method to make tree sampling estimates for germplasm conservation, and defined unique seed zones for American chestnut based on climate and genotype, and 3) determined the amount of wild adaptive diversity captured by the backcross breeding program and made recommendations for their replanting region. These results will inform the development of a breeding plan for the introgression of adaptive diversity into backcross and transgenic chestnut populations. / Doctor of Philosophy / The American chestnut tree (Castanea dentata) is a deciduous tree that largely exists in the eastern United States along the Appalachian Mountain range. Approximately 100 years, a fungal disease (Cryphonectria parasitica) decimated chestnut populations, resulting in the loss of billions of trees. The American Chestnut Foundation developed disease-resistant American chestnut backcross trees by breeding American chestnut trees with Chinese chestnut trees (Castanea mollissima). These trees will need additional breeding with wild American chestnut trees so that their offspring will have both the disease-resistant traits and the adaptations to the local environment where they will be replanted. This is important, because trees that are both disease-resistant and locally adapted will be most likely to survive and thrive in their replanting location. However, a comprehensive evaluation of the genomic basis for local adaptation in American chestnut populations is lacking. In this dissertation, I presented our findings which addressed previous gaps in knowledge regarding the population genomics of wild and backcross American chestnut populations. I 1) estimated the genomic diversity, number of unique populations, and population size changes over time in wild American chestnut; 2) revealed the genes related to local adaptation in American chestnut, developed a novel method to make tree sampling estimates for conserving wild American chestnut diversity, and defined unique seed zones (areas within the species range that have unique adaptations to environment) for American chestnut based on climate (ie, precipitation and temperature values) and genotype (DNA), and 3) determined the amount of wild genomic diversity related to local adaption captured by the backcross breeding program and made recommendations for their replanting region. These results will inform the development of a breeding plan of wild American chestnut with backcross and transgenic chestnut populations to create locally adapted and disease-resistant chestnut populations for reintroduction.

Les progrès technologiques récents nous permettent d'accéder à la variation des génomes complets ce qui nous ouvre la porte d'une meilleure compréhension de leur diversification via des approches de génomique des populations et de génomique du paysage. Ce travail de thèse se base sur l'analyse des données de génomes complets (WGS) pour caractériser la diversité génétique des petits ruminants (chèvre et moutons) et rechercher les bases génétiques d'adaptations locales.Dans un premier temps, ce travail appréhende un aspect méthodologique et examine la précision et le biais de différentes approches d'échantillonnage des génomes pour caractériser la variabilité génétique, en les comparant aux données WGS. Nous mettons en évidence un fort biais des approches classiques (i.e. puces à ADN, capture de l'exome) ainsi que des séquençages de génomes à faibles taux de couverture (1X et 2X), et nous suggérons des alternatives basées sur un échantillonnage aléatoire de marqueurs dont la densité est variable selon les objectifs d'étude (évaluation de la diversité neutre, déséquilibre de liaison, signatures de sélection). Le jeu de données produit a permis d'évaluer l'état des ressources génétiques de différentes populations domestiques (races locales marocaines, iraniennes, races industrielles) et sauvages (aegagre, mouflon asiatique). Nous relevons une très forte diversité génétique dans les populations indigènes et sauvages qui constituent des réservoirs d'allèles et peuvent jouer un rôle important pour préserver le potentiel adaptatif des petits ruminants domestiques dans un contexte de changement climatique. L'étude plus approfondie des populations de chèvres du Maroc montre une forte diversité génétique faiblement structurée géographiquement, et met en évidence des portions de génome présentant des signaux de sélection. Leur étude montre l'existence de mécanismes adaptatifs potentiellement différents selon les populations (e.g. transpiration/halètement dans l'adaptation probable à la chaleur).Enfin, nous explorons les bases génétiques de l'adaptation locale à l'environnement chez les moutons et chèvres via une approche de génomique de paysage. En scannant les génomes de 160 moutons et 161 chèvres représentant la diversité éco-climatique du Maroc, nous identifions de nombreux variants et gènes candidats qui permettent d'identifier les voies physiologiques potentiellement sous-jacentes à l'adaptation locale. En particulier, il apparait que les mécanismes respiratoires et les processus cardiaques joueraient un rôle clé dans l'adaptation à l'altitude. Les résultats suggèrent que les chèvres et moutons ont probablement développé différents mécanismes adaptatifs pour répondre aux mêmes variations environnementales. Cependant, nous identifions plusieurs cas probables de voies adaptatives communes à plusieurs espèces. Par ailleurs, nous avons caractérisé les patrons de variations du niveau de différenciation de régions chromosomiques sous sélection en fonction de l'altitude. Cela nous permet de visualiser la diversité des réponses adaptatives selon les gènes (par exemple, sélection de variants à faible et/ou haute altitude). Ainsi, ce travail pose les bases de la compréhension de certains mécanismes d'adaptation locale. / Recent technological developments allow an unprecedented access to the whole genome variation and would increase our knowledge on genome diversification using population and landscape genomics. This work is based on the analysis of Whole Genome Sequence data (WGS) with the purpose of characterising genetic diversity in small ruminants (sheep and goats) and exploring genetic bases of local adaptation.First, we addressed a methodological aspect by investigating the accuracy and possible bias in the widely used genotyping approaches to characterize genetic variation in comparison with WGS data. We highlighted strong bias in conventional approaches (SNP chips and exome capture) and also in low-coverage whole genome re-sequencing (1X and 2X), and we suggested effective solutions based on sampling panels of random markers over the genome depending the purpose of the study (assessing neutral diversity, linkage disequilibrium, selection signatures). The various datasets produced allowed assessing genetic resources in various domestic (Moroccan and Iranian indigenous breeds and industrials) and wild populations (bezoars and Asiatic mouflons). We identified a very high diversity in indigenous and wild populations. They constitute a reservoir of alleles allowing them to play a possible key role in the preservation of these species in the context of global changes. The deep study of Moroccan goats showed a high diversity that is weakly structured in geography and populations, and highlighted numerous genomic regions showing signatures of selection. These regions identified different putative adaptive mechanisms according to the population (e.g. panting/sweating to adapt to warm/desert environment).Then, we explored genetic bases of local adaptation to the environment in sheep and goats using a landscape genomics framework. We scanned genomes of 160 sheep and 161 goats representing the eco-climatic Moroccan-wide diversity. We identified numerous candidate variants and genes, which allowed for identifying physiological pathways possibly underlying local adaptation. Especially, it seems that respiration and cardiac process have key roles in the adaptation to altitude. Our results suggest dissimilar adaptive mechanisms for the same environment in sheep and goats. However, we highlighted several cases of common metabolic pathways in different species. Moreover, we characterized some patterns for the variation of genetic differentiation in some candidate genomic regions over environmental gradients. This allowed us to visualise different adaptive reaction depending genes. This work points the way towards a better understanding of some mechanisms underlying local adaptation.

Avec les changements climatiques et les perturbations humaines, de nombreuses espèces changent leurs aires de répartition. Dans ce contexte, l’identification de loci potentiellement adaptatifs chez ces populations en expansion est importante pour mieux comprendre le potentiel évolutif et la capacité d’envahissement de ces espèces. Toutefois, chez les espèces en expansion de territoire, tel le dendroctone du pin, les processus démographiques comme le surf d’allèles peuvent résulter en des patrons spatiaux de variation génétique neutre qui imitent ceux issus des processus adaptatifs. Ce phénomène gonfle le taux de faux-positif d’identification des loci adaptatifs et confond ainsi les méthodes de génomique du paysage. Dans le cadre de ce projet de maîtrise, j’ai étudié le développement des structures génétiques neutres et adaptatives lors d’une expansion de territoire. Je me suis attardé particulièrement sur l’influence de différentes conditions démographiques sur le taux de loci neutres qui imitent les patrons spatiaux de ceux adaptatifs, en menant une revue de littérature et en utilisant le système épidémique du dendroctone du pin pour paramétrer un modèle de simulations. J’ai simulé les dynamiques démographiques et génétiques des populations de dendroctones à l’aide du modèle de simulation génétique explicitement spatial CDmetaPOP. J’ai analysé les conséquences de trois facteurs sur le taux de faux-positif : 1) la capacité de dispersion; 2) le moment d’échantillonnage durant l’expansion de territoire; et 3) la force de sélection agissant sur le locus adaptatif de référence. J’ai démontré qu’une combinaison de faible capacité de dispersion, faible sélection et un échantillonnage tôt au début de l’expansion contribuent à un plus grand taux de faux-positif, alors qu’une forte capacité de dispersion conduit à de plus faibles taux de faux-positif. Lorsque les méthodes de génomique du paysage sont utilisées dans ces conditions, elles risquent d’avoir un haut taux de loci neutres identifiés comme adaptatifs et doivent donc être interprétées avec prudence. La situation démographique complexe que présente le système actuel du dendroctone (et d’autres espèces irruptives et envahissantes) rend l’identification d’allèles adaptatifs plus difficile. Les résultats de ce projet encouragent l’incorporation de ces processus démographiques dans les méthodes de génomique du paysage. / Under the actual climate changes and human perturbations global context, many species are altering their geographic range. The identification of putatively adaptive loci in those expanding populations is thus important to better understand evolutionary potential and invasiveness of these species. However, in irruptive species undergoing rapid expansion, such as the mountain pine beetle (MPB), demographic processes such as allele surfing can result in spatial patterns of neutral genetic variation that can mimic those that result from adaptive processes. This phenomenon inflates the false discovery rate of adaptive loci and thus confounds landscape genomics methods. In this thesis, I studied the development of neutral and adaptive genetic structure during a range expansion. I investigated precisely how different demographic conditions influence the rate of neutral loci mimicking the spatial patterns of adaptive loci, doing a literature review and using the mountain pine beetle outbreak system to parametrize a simulation model. I simulated the demographic and population genetic dynamics of mountain pine beetle populations undergoing range expansion using the spatially explicit, individual-based genetic model CDmetaPOP. I examined the consequences of three factors on the false discovery rate: 1) species dispersal capacity; 2) timing of sampling during the course of the expansion; and 3) the strength of selection on adaptive reference loci. I found that a combination of weak dispersal capacity, weak selection, and early sampling during expansion results in the highest rate of false positive, while strong dispersal was responsible for lower rates of false positive. Used under these conditions of dispersal capacity, strength of selection and sampling timing, landscape genomics models risk elevated false discovery rates of adaptive loci and must be interpreted cautiously. Complex demography in the current MPB system (and other irruptive and invasive species) makes identification of adaptive loci challenging. Results from this project clearly demonstrate that there is a need for further method development to include these directional demographic processes in the field of landscape genomics.

Attività antropiche e pressioni di mercato stanno rapidamente riducendo la biodiversità. Per questa ragione, conservare il patrimonio ecosistemico, tassonomico e genetico risulta fondamentale al fine di garantire potenziale adattativo alle specie, e, in ultima analisi, un futuro sostenibile per il pianeta. Al fine di minimizzare la perdita di biodiversità, numerosi metodi sono stati proposti per priorizzare ecosistemi, specie e popolazioni. Il presente lavoro di tesi fornisce in primo luogo una revisione di tali approcci, proponendo un albero decisionale volto a favorirne un corretto utilizzo. Secondariamente, la variabilità genomica neutrale del bufalo d’acqua (Bubalus bubalis L.) è investigata per mezzo di un pannello di marcatori SNP a media densità, rivelando due centri di domesticazione (India Nord-occidentale, Cina-Indocina) e possibili rotte di migrazione per gli ecotipi ‘river’ e ‘swamp’. L’adattamento locale ad East Coast Fever, patologia endemica delle popolazioni bovine in Africa Sub-sahariana, è stato inoltre studiato in bovini autoctoni Ugandesi (Bos taurus L.) combinando tecniche di modellizzazione delle nicchie ecologiche e di genomica del paesaggio. L’approccio ha portato ad indentificare PRKG1 e SLA2 come possibili geni di adattamento. I risultati sono discussi alla luce delle possibili implicazioni nella conservazione del bufalo e nella gestione delle risorse genetiche animali Ugandesi. / Biodiversity is quickly disappearing due to human impact on the biosphere, and to market pressure. Consequently, the protection of both wild and domestic species needs to become a priority in order to preserve their evolutionary potential and, ultimately, guarantee a sustainable future for coming human generations. To date, tens of methods have been proposed to prioritize biodiversity for conservation purposes. Here, an ontology for priority setting in conservation biology is provided with the aim of supporting the selection of the most opportune methodologies given specific conservation goals. Further, two case studies are presented characterizing neutral and adaptive genomic diversity in water buffalo (Bubalus bubalis L.) and indigenous Ugandan cattle (Bos taurus L.), respectively. In particular, two independent domestication centres (North-western India and Indochina) and separate migration routes are suggested for the ‘river’ and ‘swamp’ water buffalo types. In the case of indigenous Ugandan cattle, the integration of species distribution modelling and landscape genomics techniques allowed the identification of PRKG1 and SLA2 as candidate genes for local adaptation to East Coast Fever, a vector-borne disease affecting bovine populations of Sub-Saharan Africa. Results are discussed for their implications in water buffalo conservation and Ugandan cattle adaptive management.