No description available.

The primary goal of the Thoroughbred industry is to breed and train superior equine athletes capable of excelling on the racetrack. To date, research into the genetic underpinnings of athletic ability has been limited in the horse. Advances in equine genomics and the genetics of athletic performance in humans have opened up the possibility of investigating this important trait in the Thoroughbred. Initially, 46 candidate genes associated with human athletic performance were mapped in the equine genome by radiation hybrid (RH) and fluorescent in situ hybridization (FISH) mapping. RH data and later the draft equine genomic sequence allowed us to identify microsatellites adjacent to these and other candidate genes (95 in total). Additional microsatellites were added to increase genome coverage, producing a final panel of 186 markers. All the potential markers were initially screened on a pool of DNA for 16 Thoroughbreds to ensure they were polymorphic. The panel was genotyped on 162 Thoroughbreds in total; Centimorgans (cM) between microsatellites were determined with CRI-MAP. The animal?s athletic ability was estimated using career winnings loge transformed to create a linear trait; unraced animals were treated as missing data. Linkage analysis was carried out using the MERLIN program, and association analysis was carried out using the QTDT program. Appropriate thresholds for statistical significance were determined by carrying out 1000 simulated genome scans based on the structure of the original data. LOD scores above 1.54 met the criteria of statistical significance (with a 5% chance of type I error). In the actual genome scan, the marker L12.2 had the highest observed LOD score of 1.16 and p-value of 0.01 and consequently was not significant; the association analysis also did not detect significant association with performance on the track. Given the complexity of the phenotype under investigation and the modest sample size, the lack of linkage/association was not unexpected. Nevertheless, this study has contributed to the RH and FISH maps of the equine genome. Additionally, the development of the genome scanning panel for this study has provided useful information on the most informative microsatellites for linkage or association studies in the Thoroughbred.

The last century, humans have been altering almost all natural environments at an accelerating rate, including the Baltic Sea that has highly eutrophicated areas and many coastal industries such as Pulp-mills. For animals living in a habitat that changes there are basically two alternatives, either to cope with the change or become locally extinct. This thesis aims to investigate if recent anthropogenic disturbance in the Baltic Sea can affect natural populations on a genetic level through natural selection. First, we found a fine-scale genetic structure in three-spine sticklebacks (Gasterosteus aculeatus) populations along the Swedish coast (paper I), indicating limited gene-flow between populations in geographic proximity. Different genetic markers, specifically Amplified Fragment Lenght Polymorpism (AFLP, and microsatellites,  gave different results, highlighting the heterogeneous character of genomes which demonstrates that it is important to choose a genetic marker that is relevant for the question at hand. With a population genomic approach, and a multilocus genetic marker (AFLP), we detected convergent evolution in genotype composition in stickleback populations living in environments affected by pulp-mill effluent (paper II) and in highly eutrophicated environments (paper III), compared to adjacent reference populations. We found loci, in both studies (paper II, III), that were different from a neutral distribution and thus probably under divergent selection for the habitat differences investigated. The selective effect from pulp-mill effluents were more pronounced, but the two different habitats had mutual characters (AFLP loci). In paper IV, we converted five anonymous AFLP loci to sequenced markers and aligned them to the stickleback genome. Four out of five loci aligned within, or close to, coding regions on chromosome I, chromosome VIII, chromosome XIX and chromosome XX. One of the loci, located on chromosome VIII and identified as under divergent selection in both paper II and III, has been identified in other studies as to be under selection for fresh water adaptation, including Baltic Sea stickleback populations. In conclusion, anthropogenic alterations of natural environments can have evolutionary consequences, probably adaptive, for the animals living there and the evolutionary response exhibited by natural populations can be very fast. / <p>At the time of the doctoral defense, the following papers were unpublished and had a status as follows: Paper 1: Manuscript. Paper 3: Manuscript. Paper 4: Manuscript.</p>

Domestication is the process by which animals become adapted to the environment provided by humans. The process of domestication has let to a number of correlated behavioural, morphological and physiological changes among many domesticated animal species. An example is the changes of plumage colour in the chicken. Plumage colour is one of the most readily observable traits that make distinction between breeds as well as between strains within a breed. Understanding the genetic architecture of pigmentation traits or indeed any trait is always a great challenge in evolutionary biology. The main aim of this study was to map quantitative trait loci (QTLs) affecting the red and metallic green coloration in the chicken plumage. In this study, a total of 572 F8 intercross chickens between Red Junglefowl and White Leghorn were used. Phenotypic measurements were done using a combination of digital photography and photography manipulating software. Moreover, all birds were genotyped with 657 molecular markers, covering 30 autosomes. The total map distance covered was 11228 cM and the average interval distance was 17 cM. In this analysis, a total of six QTLs (4 for red and 2 for metallic green colour) were detected on four different chromosomes: 2, 3 11 and 14. For red colour, the most significant QTL was detected on chromosome 2 at 165 cM. An additional QTL was also detected on the same chromosome at 540 cM. Two more QTLs were detected on chromosomes 11 and 14 at 24 and 203 cM respectively. Additionally, two epistatic pairs of QTLs were also detected. The identified four QTLs together can explain approximately 36% of the phenotypic variance in this trait. In addition, for metallic green colour, one significant and one suggestive QTLs were detected on chromosomes 2 and 3 at 399 and 247 cM respectively. Moreover, significant epistatic interactions between these two QTLs were detected. Furthermore, these two QTLs together can explain approximately 24% of the phenotypic variance in this trait. These findings suggest that the expression of pigmentation in the chicken plumage is highly influenced by both the epistatic actions and pleiotropic effects of different QTLs located on different chromosomes.

Degenerative suspensory ligament desmitis (DSLD) is a debilitating disease of connective tissues seen in many breeds but has become prevalent in the Peruvian Pasohorse. DSLD is believed to be a genetic disorder caused by one primary founder and most likely has a recessive mode of inheritance although a dominant or co-dominant mode of inheritance has not been ruled out. A genome scan using 259 microsatellite markers was used to test for linkage disequilibrium between one or more markers and DSLD. Two groups of Peruvian Pasohorses were selected from one population including the US and Canada. The only difference between the two groups of horses besides the size of the two groups was the presence of DSLD in the affected group and the absence of DSLD in the unaffected group. It was assumed that differences seen between the two groups in homozygosity and or common allele frequency could be an indication of linkage to DSLD. As a connective tissue disorder, there were a large number of candidate genes forDSLD to consider, yet no identical human or animal model exists. The genome scan identified five chromosomal regions where statistically significant differences were seen between affected and unaffected sample populations that could be indications of linkage to DSLD. Those chromosomes were: ECA 6, 7, 11, 14, and 26. Sequencing of a portion of the G domain in the Chondroitin Sulfate Proteoglycan2 (CSPG2) gene has mostly ruled out that segment of chromosome 14 as having linkage to DSLD. Further research needs to be conducted in the regions of ECA 6,7,11 and 26 where statistically significant differences were seen between the affected and unaffected groups, especially on ECA 6 and 11 since possible candidate genes are located in those regions based on the human comparative map.

Melampsora larici-populina est un champignon pathogène responsable de la rouille foliaire sur les peupliers, causant de graves dommages dans les plantations du monde entier. Presque toutes les résistances des peupliers déployées en France ont été contournées et l'événement majeur est survenu en 1994 avec le contournement de la résistance R7 largement déployée en populiculture. Dans le but d'identifier des gènes candidats liés à la pathogénicité, j'ai mené une étude de génomique comparative basée sur le séquençage de 15 isolats. Cette analyse a mis à jour des patrons de polymorphisme corrélés à la distribution des virulences au sein des isolats tout en révélant la nécessité d'une étude populationnelle. Pour se faire, une étude de génétique des population basée sur le génotypage de 600 isolats de M. larici-populina échantillonnés de 1992 à 2012 a été conduite. Cette analyse m'a permis de décrire l'histoire démographique des populations de M. larici-populina et de documenter l'impact majeur du contournement de la R7 sur la structure des populations. Enfin, j'ai mené une analyse de génomique des populations afin d'obtenir un scenario démographique décrivant les liens historiques entre les populations et d'identifier les régions sous sélection. Cette analyse est basée sur le séquençage en Illumina de 86 isolats répartis en quatre populations clés mises en évidence par l'analyse de la structure génétique des populations. Plus de 1 000 000 positions polymorphes ont été identifiées. Un scenario fiable a été identifié par ABC à partir duquel les enveloppes de confiance des indices de génétique des populations ont été mesurées. L'analyse du génome scan qui a été réalisée sur les 86 génomes en utilisant ces mêmes indices a révélée 20 régions génomiques contenant 14 gènes candidats potentiellement impliquées dans le contournement de la R7. / Melampsora larici-populina is a pathogenic fungus responsible of poplar leaf rust, causing severe damages in plantations worldwide. Almost all poplar resistances deployed so far in France have been overcome and a major event that occurred in 1994 with the breakdown of resistance R7 mostly used in poplar cultivation. In order to identify candidate genes linked to pathogenicity, I conducted a comparative genomics study based on the sequencing of 15 isolates. This analysis revealed polymorphism patterns correlated to the distribution of virulences among isolates while the necessity of a population genetics study. I then analyzed the genetic structure of a comprehensive collection of 600 isolates of M. larici-populina sampled from 1992 to 2012. This analysis demonstrated the major impact of the R7 breakdown on populations. Finally, I conducted a population genomics analysis to obtain a demographic scenario describing the historical links between populations and to identify genomic regions under selection. This analysis is based on the Illumina sequencing of 86 isolates in four key populations identified by the population genetic analysis. Over 1,000,000 polymorphic positions were identified. The best demographic scenario was assessed using Approximate Bayesian Computation algorithms based on coalescent simulations. Using this demographic scenario, I computed the confidence interval of several population genetic indices. This genome scan analysis was performed on the 86 genomes using this same indices and revealed 20 genomic regions containing 14 genes potentially involving in the resistance 7 breakdown

[ES] La selección divergente puede cambiar las frecuencias de los marcadores genéticos en direcciones opuestas, produciéndose frecuencias alélicas intermedias en estos marcadores cuando ambas líneas divergentes son consideradas conjuntamente en los análisis genéticos. Por lo tanto, los experimentos de selección divergente aumentan el poder de detección para estudios de asociación de genoma completo (GWAS) y para estudios de escaneo genómico por medio de métodos de huellas de selección. GWASs bayesianos, utilizando el modelo Bayes B, se implementaron para analizar datos genómicos de los caracteres de tamaño de camada del experimento de capacidad uterina con 181 hembras. Las asociaciones fueron evaluadas calculando los factores de Bayes para cada SNP, y calculando los porcentajes de la varianza genómica para cada ventana no solapada de 1-Mb. Los GWASs descubrieron SNPs asociados con el número total de gazapos al parto y los embriones implantados. Además, se revelaron regiones genómicas relevantes para el número total de gazapos al parto (1 región), el número de nacidos vivos (1 región), los embriones implantados (3 regiones) y la tasa de ovulación (5 regiones). Los porcentajes de varianza genómica que explicaban los anteriores caracteres de tamaño de camada fueron 39,48%, 10,36%, 37,21% y 3,95%, respectivamente, en un modelo que excluye el efecto línea; y 7.36%, 1.27%, 15.87% y 3.95%, respectivamente, en un modelo con el efecto línea. La región genómica localizada en el cromosoma del conejo (OCU) 17 en 70.0 - 73.3 Mb se consideró como un nuevo locus de carácter cuantitativo (QTL) asociado a caracteres reproductivos en conejos, ya que esta región fue encontrada solapada para el número total de gazapos al parto, el número de nacidos vivos y los embriones implantados. El gen de la proteína morfogenética ósea 4, BMP4, es el principal gen candidato prometedor dentro del nuevo QTL. Una combinación de GWASs fueron implementados para analizar los datos genómicos del experimento de la grasa intramuscular con 480 conejos. Los métodos de GWASs incluyeron un método bayesiano, modelo Bayes B; y un método frecuentista, regresiones de marcadores únicos con los datos ajustados por el parentesco genómico. Este estudio reveló cuatro regiones genómicas relevantes en OCU1 (1 región), OCU8 (2 regiones) y OCU13 (1 región) asociadas con la grasa intramuscular. La región asociada más importante estaba en OCU8 en 24.59 - 26.95 Mb, y explicó el 7.34% de la varianza genómica. El bajo porcentaje explicado por las principales regiones genómicas relevantes indica un gran componente poligénico para la grasa intramuscular. Los análisis funcionales recuperaron genes vinculados con las rutas y funciones de los metabolismos de energía, carbohidratos y lípidos. Además, se realizó un estudio de escaneo genómico usando conejos del experimento de selección divergente para grasa intramuscular, y usando tres métodos de firmas de selección: índice de fijación de Wright (Fst), coeficiente de verosimilitud compuesto entre poblaciones (XP-CLR) y extensión de homocigosidad de los haplotipos entre poblaciones (XP-EHH). Los resultados mostraron múltiples huellas de selección en todo el genoma del conejo. Ninguna de estas huellas de selección concuerda con las regiones genómicas asociadas con la grasa intramuscular, provenientes de los resultados de los GWASs. En síntesis, los resultados de ambos experimentos, GWASs y el estudio de escaneo genómico, sugieren que la arquitectura genómica de la grasa intramuscular en el conejo parece ser altamente poligénica y sus variantes causales serían apenas detectables. Este estudio demuestra que la detección de variantes causales y marcadores genéticos asociados depende de las hipotéticas arquitecturas genómicas de los caracteres, independientemente de las exitosas respuestas logradas en los dos experimentos de selección divergente. Hasta la fecha, estos hallaz / [CAT] La selecció divergent pot alterar les freqüències dels marcadors genètics en direccions oposades, donant lloc a freqüències al·lèliques intermèdies quan les dos línies divergents es consideren conjuntament en els anàlisis genètics. Per tant, els experiments de selecció divergents augmenten el poder de detecció en estudis d'associació de genoma ampli (GWAS) i en estudis d'exploració genòmica a través de mètodes de signatures de selecció. GWASs bayesians, utilitzant el model Bayes B, es van implementar per a analitzar dades genòmiques de caràcters de grandària de ventrada de l'experiment de capacitat uterina amb 181 conilles femelles. Les associacions es van provar calculant els factors de Bayes per a cada SNP, i calculant els percentatges de la variància genòmica per a cada finestra no superposada d'1-Mb. Els GWASs van descobrir SNPs associats amb el número total de llorigons al part i els embrions implantats. A més, es van revelar regions genòmiques rellevants per al número total de llorigons al part (1 regió), el número de nascuts vius (1 regió), els embrions implantats (3 regions) i la taxa d'ovulació (5 regions). Els percentatges de variància genòmica que explicaven els anteriors caràcters de grandària de ventrada van ser 39,48%, 10,36%, 37,21% i 3,95%, respectivament, sota un model que exclou l'efecte de línia; i 7.36%, 1.27%, 15.87% i 3.95%, respectivament, sota un model amb efecte de línia. La regió genòmica situada en el cromosoma del conill (OCU) 17 en 70.0 - 73.3 Mb es va considerar com un nou locus de caràcters quantitatius (QTL) associat a caràcters reproductius en conills, ja que aquesta regió es va superposar per al número total de llorigons al part, el número de nascuts vius i els embrions implantats. El gen de la proteïna morfogenètica òssia 4, BMP4, és el principal gen candidat prometedor dins del nou QTL. Una combinació de GWASs es van implementar per a analitzar les dades genòmiques de l'experiment del greix intramuscular amb 480 conills. Els mètodes GWASs van incloure un mètode bayesià, model Bayes B; i un mètode frecuentista, regressions de marcadors únics amb les dades ajustades pel parentiu genòmico. Aquest estudi va revelar quatre regions genòmiques rellevants en OCU1 (1 regió), OCU8 (2 regions) i OCU13 (1 regió) associades amb el greix intramuscular. La regió associada més important estava en OCU8 en 24.59 - 26.95 Mb, i va explicar el 7.34% de la variància genòmica. El baix percentatge explicat per les principals regions genòmiques rellevants indica un gran component poligènic per al greix intramuscular. Els anàlisis funcionals van recuperar gens relacionats amb les rutes i la funció d'energia, metabolismes de carbohidrats i lípids. A més, es va realitzar un estudi d'exploració del genoma usant conills de l'experiment de selecció divergent per a greix intramuscular, i usant tres mètodes de signatures de selecció: índex de fixació de Wright (Fst), coeficient de versemblança compost entre poblacions (XP-CLR) i extensió de homocigosidad dels haplotipos entre poblacions (XP-EHH). Els resultats van mostrar múltiples petjades de selecció en tot el genoma del conill. Cap d'aquestes petjades de selecció concorda amb les regions genòmiques associades a partir dels resultats dels GWASs. En síntesi, els resultats dels dos experiments, GWASs i estudi d'exploració del genoma, suggereixen que l'arquitectura genòmica del greix intramuscular en el conill sembla ser altament poligènica i les seues variants causals serien a penes detectables. Aquest estudi demostra que la detecció de variants causals i marcadors genètics associats depèn de les hipotètiques arquitectures genòmiques dels caràcters, independentment de les respostes reeixides en els dos experiments de selecció divergents. Fins ara, aquestes troballes no tindrien implicacions valuoses per als programes de cria de conills. / [EN] Divergent selection can alter frequencies of genetic markers in opposite directions, leading to intermediate allelic frequencies when both divergent lines are jointly considered in the genetic analyses. Therefore, divergent selection experiments increase the detection power for genome wide association studies (GWAS) and for genomic scan studies through methods of selection signatures. Bayesian GWASs using Bayes B model was used to analyse genomic data of litter size traits of the uterine capacity experiment with 181 does. The associations were tested by computing Bayes factors for each SNP, and by computing percentages of the genomic variance for each 1-Mb non-overlapping window. The GWASs uncovered SNPs associated with total number born and implanted embryos. Moreover, relevant genomic regions were revealed for total number born (1 region), number born alive (1 region), implanted embryos (3 regions), and ovulation rate (5 regions). The percentages of genomic variance that accounted for these litter size traits were 39,48%, 10.36%, 37.21%, and 3.95%, respectively, under a model excluding line effect; and 7.36%, 1.27%, 15.87%, and 3.95%, respectively, under a model with line effect. The genomic region located on the rabbit chromosome (OCU) 17 in 70.0 - 73.3 Mb was deemed as a novel quantitative trait locus (QTL) of reproductive traits in rabbits, since this region was found overlapped for total number born, number born alive and implanted embryos. Bone morphogenetic protein 4 gene, BMP4, is the main promising candidate gene within the novel QTL. A combination of GWASs were performed for analysing the genomic data of the intramuscular fat experiment with 480 rabbits. The GWAS methods included a Bayesian method, Bayes B model; and a frequentist method, single marker regressions with the data adjusted by genomic relatedness. This study revealed four relevant genomic regions in OCU1 (1 region), OCU8 (2 regions) and OCU13 (1 region) associated with intramuscular fat. The most important associated region was on OCU8 in 24.59 - 26.95 Mb, and accounted for 7.34% of the genomic variance. The low percentage explained by the main relevant genomic regions indicates a large polygenic component for intramuscular fat. Functional analyses retrieved genes linked to pathways and function of energy, carbohydrate and lipid metabolisms. In addition, a genome scan study was performed using rabbits from the divergent selection experiment for intramuscular fat, and using three methods of selection signatures: Wright's fixation index (Fst), cross population composite likelihood ratio (XP-CLR) and cross population extended haplotype homozygosity (XP-EHH). The results showed multiple selection signatures across the rabbit genome. None of these selection signatures agreed with the associated genomic regions from GWAS findings. In synthesis, the results of both experiments, GWAS and genome scan study, suggest that the genomic architecture of intramuscular fat in rabbit seems to be highly polygenic and their causative variants would be hardly detectable. This study demonstrates that detection of causative variants and associated genetic markers depends on the hypothetical genomic architectures of traits, regardless of the successful responses attained in the two divergent selection experiments. Hitherto, these findings would not have worthwhile implications for the rabbit breeding programs. / Sosa Madrid, BS. (2020). Genomic analysis of divergently selected experimental lines in rabbit [Tesis doctoral no publicada]. Universitat Politècnica de València. https://doi.org/10.4995/Thesis/10251/141376 / TESIS

The seeming lack of barriers to gene flow in the northwest Atlantic ocean has led to the general assumption that the population of the American lobster (Homarus americanus) is largely panmictic. However, morphological and genetic data presented in this dissertation suggest that lobster populations are less homogenous than once believed with potential for behavioral barriers to mating and selection of locally adaptive traits. Additionally, both long-term fishing pressures and the recent spread of a destructive epizootic shell disease may have impacted population structure. We developed a novel photographic technique to rapidly collect accurate morphological data with the ability to maintain a database of images for the purposes of re-sampling and testing additional hypotheses. During this study, we found significant morphometric differences between samples of lobsters from collection sites as close as 25km apart. Morphological differences may have originated due to differential selection or plastic responses to environmental variation. To analyze population genetic structure, I surveyed genetic variation using RADseq. Analysis of 1614 putatively neutral SNPs found little genetic difference (Average F¬ST=0.00137) between sample sites suggesting a high level of gene flow between regions. Several additional markers appeared to be under divergent selection between sample sites. A genome scan analysis of both neutral SNPs and SNPs under selection found several selected SNPs associated with principal components of morphological characters. A subsequent BLAST analysis identified a number of the selected SNPs lying in the H. americanus transcriptome, suggesting functional importance. Further experimentation is required to quantify the impacts of plasticity or local adaption in the origin of morphological differences between lobster populations, although the significant differences identified in this research are likely due to a combination of the two. The overarching conclusion is that lobster populations are, in fact, more differentiated than previously predicted and, as such, the findings presented here may have significant management implications.

Les bases génétiques et l'histoire de la différenciation adaptative ont été étudiées chez les moules du complexe d'espèces Mytilus edulis qui représente un modèle d'étude intéressant pour mieux comprendre comment se propage et se distribue la différenciation adaptative en population structurée. Grâce à la technique AFLP, une approche de génomique des populations (« scan génomique ») a été utilisée pour mesurer la différenciation entre des populations isolées sur des échelles de temps et d'espaces contrastées. Notre objectif était de vérifier si la différenciation génétique n'avait pas une origine plus complexe qu'habituellement proposé. Trois parties s'articulent autour de cette question centrale. La première s'intéresse à la différenciation entre l'Atlantique et la Méditerranée chez l'espèce M. galloprovincialis et a montré que la structure génétique était la conséquence d'un différentiel d'introgression avec l'espèce sœur M. edulis. Dans la deuxième partie de ce travail nous avons mis en évidence qu'une sélection directe, soit balancée soit intermittente, sur un polymorphisme pré-existant expliquait le niveau de différenciation anormalement élevé d'un gène de l'immunité entre populations de la côte européenne chez M. edulis. Enfin, la troisième partie s'est intéressée à revisiter un exemple classique de la littérature de la génétique des populations: le cas des M. edulis du détroit de Long Island et a permis de suggérer que la structure observée à très petite échelle spatiale correspondait probablement à un contact secondaire entre des moules européennes introduites et les moules américaines. D'une manière générale nos résultats démontrent que, quelque soit l'échelle à laquelle nous nous plaçons, la différentiation génétique tire son origine d'une histoire souvent plus complexe qu'attendu. / The genetic basis and history of adaptive differentiation were studied in the species complex M. edulis which is an interesting model system to understand how adaptive differentiation spreads and structure itself in subdivided populations. Using the AFLP technique, a genome scan approach was undertaken to measure differentiation between populations on contrasted spacial and temporal scales. Our objective was to verify wether the origin of genetic differentiation could be more complex than anticipated. This question was addressed in three chapters. The first one focuses on the differentiation between populations of the Atlantic Ocean and Mediterranean Sea in M. galloprovincialis. Our results show that the genetic structure was the result of differential introgression with the sister hybridizing species M. edulis. In the second chapter of this work we demonstrated that direct selection on a pre-existing polymorphism explained the unusually high level of differentiation at a defensin locus between populations of the European coast in M. edulis. Finally, in the third chapter we revisited a classic example of the population genetics literature: the aminopeptidase cline in M. edulis populations of the Long Island Sound. We obtained evidence that the genetic differentiation observed at a very fine spatial scale in the sound was the consequence of a secondary contact between introduced mussels from Europe and indigenous American mussels. Whatever the spatio-temporal scale at which we analyzed genetic differentiation, its origin proved to originate from an unsuspectedly long and complex history.

Master of Science / Department of Plant Pathology / Eduard D. Akhunov / Loretta C. Johnson / Big bluestem (Andropogon gerardii Vitman) is an ecologically dominant grass of the North American grasslands with precipitation-dependent productivity. However, climatic predictions for big bluestem’s dominant range in the Great Plains include increased periods of drought. The main objectives of this research were to determine the extent of neutral and non-neutral genetic differentiation and diversity among putative big bluestem ecotypes using amplified fragment length polymorphism (AFLP) markers. This is the first study of both neutral and non-neutral genetic diversity of big bluestem which also includes source populations of well-described ecotypes studied in reciprocal common gardens. A total of 378 plants were genotyped from 11 source prairies, originating from one of three ecoregions (Central Kansas, Eastern Kansas, and Illinois). Using two AFLP primer sets, 387 polymorphic markers (error rate 9.18%) were found. Un-rooted neighbor joining tree and principle-component analyses showed continuous genetic differentiation between Kansas and Illinois putative ecotypes, with genetic overlap occurring between Kansas ecotypes. Analysis of molecular variance showed high diversity within-prairie sites (80%) relative to across-prairies (11%), and across- ecoregions (9%) (p<0.001). Within-prairie genetic diversity levels were similar among ecoregions (84-92%), with the highest genetic variation maintained in Illinois prairies (92%). Population structure analyses supported K=6 genetic clusters across the environmental gradient, with Kansas prairies belonging to three main genetic groups, and Illinois prairies having largely divergent allele frequencies from Kansas prairies. Interestingly, BAYESCAN analysis of the three putative ecotypes identified eight F[subscript]ST-outlier AFLP loci under potential diversifying selection. Frequency patterns of loci under diversifying selection were further linked to geo-environmental descriptors including precipitation, temperature severity, diurnal temperature variation, prairie location, and elevation. The observed allele frequency divergence between Kansas and Illinois ecotypes suggests tallgrass restorations should consider possible maladaptation of non-local ecotypes and genetic swamping. However, high within-prairie genetic variation may help individual big bluestem populations withstand climatic variability.