Orientadores: Anete Pereira de Souza, Antonio Augusto Franco Garcia / Tese (doutorado) - Universidade Estadual de Campinas, Instituto de Economia / Made available in DSpace on 2018-08-20T23:12:17Z (GMT). No. of bitstreams: 1 Souza_LiviaMourade_D.pdf: 2370923 bytes, checksum: b0c800b4425fab2c0c9fda9cbbb4798d (MD5) Previous issue date: 2012 / Resumo: A seringueira é uma espécie de cruzamento misto e com um longo ciclo de vida, o que dificulta a geração de linhagens endogâmicas e, por consequência, a construção e integração de mapas de ligação usando as metodologias convencionais. A exemplo do que tem sido feito em outras espécies vegetais que apresentam limitações para a obtenção de linhagens endogâmicas, os mapas genéticos existentes para seringueira foram construídos utilizando populações F1 (com diferentes tipos de segregação). A integração dos mapas obtidos para cada um dos genitores é possível com base em marcadores bi-parental, onde ambos os genitores são heterozigóticos podendo segregar nas proporções 1:1:1:1, 1:2:1 e 3:1 na progênie F1. Além disso, o uso de marcadores codominantes e multialélicos, como os microssatélites, acrescentam muito à construção do mapa, uma vez que permite a obtenção de estimativas de frequência de recombinação e da fase de ligação com um menor viés. Os microssatélites são marcadores moleculares consagrados, sendo eles a ferramenta de escolha no estudo de diversos organismos pela simplicidade de uso e de análise. Entretanto, para espécies que ainda não têm grande parte de seu genoma sequenciado, a obtenção de marcadores desse tipo passa pela necessidade de desenvolvimento via bibliotecas genômicas. Esse trabalho objetivou desenvolver, caracterizar e mapear microssatélites em Hevea brasiliensis para possibilitar estudos genético-moleculares da variação morfológica, identificação de regiões genômicas associadas a fenótipos de interesse, bem como análises genético-populacionais. Uma biblioteca genômica enriquecida em motivos repetitivos foi construída, a partir da qual foram desenvolvidos os microssatélites utilizados nesse trabalho. A utilização dos microssatélites de Hevea brasiliensis em amplificações heterólogas envolvendo seis espécies do gênero Hevea resultou em aproximadamente 98% de sucesso nas amplificações, sugerindo a existência de um complexo formado pelas diferentes espécies do gênero. Um mapa genético-molecular foi construído utilizando-se 284 marcadores microssatélites na análise de uma população segregante F1 com 270 indivíduos, a partir do cruzamento entre os genitores PB217 e PR255. Por meio da utilização do programa ONEMAP foi possível a construção de um mapa de ligação integrado que revelou 2840 cM de extensão, distribuídos em 23 grupos de ligação. O mapeamento de QTLs realizado utilizando metodologia de mapeamento por intervalo composto (CIM) detectou 24 QTLs para altura e circunferência das plantas entre as estações de verão e inverno. Este trabalho é pioneiro na construção de um mapa integrado para seringueira pelo fato dele ter sido construído unicamente com marcadores microssatélites, os quais são altamente informativos. Além disso, é também o primeiro estudo envolvendo análise de QTL para características relacionadas ao crescimento de plantas em seringueira, bem como é inédita a aplicação da metodologia CIM para população F1 segregante / Abstract: The rubber tree is an mixed crossing species with a long life cycle, which makes it difficult for the generation of inbred lines, and, therefore, the construction and integration of linkage maps by using conventional methodologies. Similar to what has been reported from other plant species that have limitations to obtain inbred lines, the genetic maps for rubber tree have been constructed by using F1 populations (with different types of segregation). The integration of maps obtained for each of the genitors is possible based on bi-parental markers, where both genitors are heterozygous, being able to segregate in ratios 1:1:1:1, 1:2:1 and 3:1 in the F1 progeny. Moreover, the use of co-dominant and multi-allele markers, such as microsatellites, adds a lot to the construction of maps, since it allows the obtainment of estimates of recombination frequency and linkage phase with less bias. The microsatellites are established molecular markers which are the tool of choice in the study of various organisms for their simplicity of use and analysis. Nevertheless, their application in species whose genomes have not yet been sequenced requires a prior development phase. The present study intended to develop, characterize and map microsatellites for the species Hevea brasiliensis so that initiatives concerning morphological variation, identification of genomic regions linked to phenotypes of interest, as well as population genetic analysis. A repetitive DNA-enriched library was constructed from which it was developed the microsatellites used in this work. The use of Hevea brasiliensis microsatellites for heterologous amplification in other six Hevea species was done with approximately 98% of success ratio, suggesting the existence of a complex formed by different species. A molecular genetic map was constructed using 284 microsatellite markers in the analysis of an F1 segregating population with 270 individuals from a cross between the parents PB217 and PR255. By using the program ONEMAP it was possible the construction of an integrated linkage map that revealed 2840 cm in length, divided into 23 linkage groups. The QTL mapping methodology was performed using composite interval mapping (CIM) and detected 24 QTLs for plant height and circumference between summer and winter. This work is a pioneer in building an integrated map for rubber because it was built only with microsatellite markers, which are highly informative. Furthermore, it is also the first one involving the analysis of QTL for characteristics related to the growth of rubber tree plants, as well as novel application of the CIM methodology for this type of population / Doutorado / Genetica Vegetal e Melhoramento / Doutor em Genetica e Biologia Molecular

Doctor of Philosophy / Genetics Interdepartmental Program / Allan K. Fritz / Wheat diseases cause significant economic losses every year. To ensure global food security, newly released cultivars must possess increased levels of broadly-effective resistance against wheat pathogens, acceptable end-use quality, and high yield potential. Genetic host resistance stands out from other management strategies as the most viable option for controlling diseases. New genotyping platforms allow whole genome marker discovery at a relatively low cost, favoring the identification of novel loci underlying traits of interest. The work presented here describes genomic approaches for mapping and predicting the resistance to Fusarium head blight (FHB) and wheat rusts. The first study used biparental mapping to identify quantitative trait loci (QTL) associated with Fusarium head blight (FHB) resistance. A doubled haploid population (DH) was originated from a cross of Everest and WB-Cedar, which are widely grown wheat cultivars in Kansas with moderately resistant and moderately susceptible reactions to FHB, respectively. We confirmed that neither of the parents carry known large-effect QTLs, suggesting that FHB resistance is native. Eight small-effect QTLs were identified as associated with multiple mechanisms of FHB resistance. All QTLs had additive effects, providing significant improvements in levels of resistance when they were found in combinations within DH lines. In the second study, a genome-wide association mapping (GWAS) and genomic selection (GS) models were applied for FHB resistance in a panel of 962 elite lines from the K-State Wheat Breeding Program. Significant single nucleotide polymorphisms (SNPs) associated with the percentage of symptomatic spikelets were identified but not reproducible across breeding panels tested in each year. Accuracy of predictions ranged from 0.25 to 0.51 depending on GS model, indicating that it can be a useful tool to increase levels of FHB resistance. GWAS and GS approaches were also applied to a historical dataset to identify loci underlying resistance to leaf and stem rust at seedling stage in a panel of elite winter wheat lines. Infection types of multiple races of wheat rusts from the last sixteen years of the Southern Regional Performance Nursery (SRPN) were used in this study. A total of 533 elite lines originating from several breeding programs were tested in the SRPN during this period of time. GWAS identified significant SNP-trait associations for wheat rusts, confirming the effectiveness of already known genes and revealing potentially novel loci associated with resistance.

This dissertation contains three research chapters conducted on Sclerotinia stem rot (SSR) of canola (Brassica napus L.). This disease is caused by the fungus Sclerotinia sclerotiorum and is considered endemic in canola-producing areas of North Dakota. The first research chapter presents results of a study that evaluated the role of eight phenotyping scoring systems and nine variant calling and filtering methods in detection of QTL associated with response to SSR. The study, conducted on two doubled-haploid mapping populations, showed that using multiple phenotypic data sets derived from lesion length and plant mortality and imputing missing genotypic data increased the number of QTL detected without negatively affecting the effect (R2) of QTL. Nineteen QTL were detected on chromosomes A02, A07, A09, C01, and C03 in this study. The second research chapter presents results of a work that assessed the role of temperature regimes and wetness duration on S. sclerotiorum ascospore germination and ascosporic infection efficiency. This study showed that optimum ascospore germination occurred at 21 °C while it significantly decreased at 10 and 30 °C. Infection efficacy experiments indicated that extreme temperatures and interrupting wet periods were detrimental for the disease development. A logistic regression model with 75% accuracy was developed for the disease perdition. The third research chapter presents results of a study that evaluated the role of temperature on mycelial growth of 19 S. sclerotiorum isolates collected from different geographical regions and on SSR development on plant introduction (PI) lines with different levels of resistance. Mycelial growth and disease development peaked at 25 °C. While lesion expansion on resistant cultivars and the susceptible check was negatively affected at 30 °C, the disease developed significantly on the PI with a high level of susceptibility. Results of these studies provide insights into integrated management strategies of SSR.

Rice is one of the most important cereal crops for the world population. Flowering time and seed development of rice are directly related to plant regional and ecological adaptions, and productive yield. In this dissertation, to gain knowledge of seed development in rice, the status of post-translational modifications (PTMs) in developing rice seeds was investigated. Numerous modified lysine sites in developing rice seeds were identified utilizing antibody-based affinity enrichment approaches and nano-HPLC/MS/MS analyses of acetylated, succinylated, crotonylated and 2-hydroxyisobutyrylated peptides. Functional annotation analyses indicated that a wide variety of vital biological processes were targeted by lysine PTMs. A number of modified histone and non-histone proteins were found to harbor multiple PTMs, and our findings showed that many modified histone sites were conserved across plant, human, and animal systems. Comprehensive analyses of lysine modification sites illustrated that the sites were highly sequence-specific for distinct motifs. Overall, this study provides a systematic analysis of lysine PTM proteome in plants, which will serve as the basis for future investigations of the regulatory mechanisms and functions of lysine PTMs. The mechanisms of flowering time variances in response to different photoperiods were further studied in the rice mutant, HSS. QTL-seq analysis identified a major effect on chromosome 6 responsible for the phenotypic divergence between Nipponbare (wild-type) and HSS rice. Sequence and mRNA expression analyses confirmed that allelic variants of Hd1 make HSS plants less sensitive to photoperiod by altering expression level of Hd3a. Diurnal expression pattern analyses revealed that DTH8 transcripts were largely affected by Hd1 expression level in both LD and SD. This result suggested that Hd1 may able to regulate DTH8 and DTH8-Hd1 complex abundance in response to day length in rice flowering time regulation. In addition, we discussed the functions of PTMs in flowering time regulation in rice.

No description available.

Peanut (Arachis hypogaea L.) is second major legume crop grown after soybean in the United States, and its productivity is often limited by drought stress. Drought negatively impacts the yield and quality of peanut. Drought stress in peanut causes an annual loss of approximately $520 million in the United States. Improving peanut yield under water deficit conditions is crucial for peanut growers to maintain their profitability in the market. To achieve this, it is essential to either breed or adopt already available drought tolerant cultivars that can produce higher yield under water deficit conditions. Therefore, the objectives of this research were to (1) evaluate five commercially available virginia and runner type peanut cultivars for pod yield stability using multilocation trials by studying G x E interaction across 13 environments including year, location, and irrigation regime. Linn and Binns, AMMI, Shukla, Wricke's, Finlay and Wilkinson stability models were used to determine pod yield stability. Bailey and Sullivan showed higher stability and adaptability across all stability indices whereas Wynne and TUFRunner presented high mean productivity with lesser stability across environments reflecting specific adaptation to just a few environments. Bailey and Sullivan are recommended for sustainable production across the growing region of Virginia and Carolinas. The second objective (2) was identification of drought tolerance related quantitative trait loci (QTL) and genetic markers to facilitate the development of drought tolerant cultivars. Three diverse recombinant inbred line (RIL) populations, derived from crossing lines N05006 x N04074FCT (Pop-1), line N05006 x Phillips, an old virginia-type cultivar (Pop-2), and lines N08086olJCT x PI 585005 (Pop-3) were phenotyped for the Normalized Difference Vegetation Index (NDVI), Canopy Temperature Depression (CTD), SPAD-meter relative chlorophyll content of the leaves (SPAD) and wilting for QTL mapping. Mapping identified 27 minor QTL on eight chromosomes for all physiological characteristics, i.e NDVI, CTD, SPAD and wilting, with logarithmic of odds values ranging from 2.5 to 38.5 and the phenotypic variance explained by these traits from 1.04 to 11.46 %. There were 4 loci on chromosome 2 associated with NDVI in Pop-1 and Pop-3, explaining 1.8 to 10.38% of the phenotypic variation. These genomic regions may be important resources in peanut breeding programs to improve drought tolerance. Further research is needed to increase the marker density in order to fine map the identified QTL and validate markers linked with these regions. / Doctor of Philosophy / Peanut is a multi million-dollar industry in the United States, but water limitations have a detrimental impact on yield, quality, and grower income. Drought along with aflatoxin contamination are two major challenges faced by U.S peanut industry. Annual losses to peanut caused by drought are around $520 million in the United States. Irrigation can alleviate water shortage in drought prone regions, but around 65% of peanut production in U.S is under rainfed condition, meaning that only rainfall can satisfy peanut crop water requirements. The most feasible and economical solution to peanut growers under these circumstances is to adopt drought tolerant varieties. In this research, our goal was to facilitate breeding drought tolerant cultivars through identification of molecular markers associated with drought tolerance and to identify already available drought tolerant peanut cultivars that could be a game changer for the producers. Therefore, the objectives of my research were to (1) evaluate in multiple environments five commercially available virginia and runner type peanut cultivars for pod yield stability and grade factors. In this study, we considered 13 environments, including 4 years, 4 locations, and 2 water regime. Statistical tools including Linn and Binns, AMMI, Shukla, Wricke's, and Finlay and Wilkinson were used to determine pod yield stability. These stability indices showed that Bailey and Sullivan are more stable and adaptable across different locations in terms of yield, whereas Wynne and TUFRunner presented high mean yield with lesser stability showing specific adaptation to only few environments. Based on stability analysis, Bailey and Sullivan are recommended for sustainable production across different growing region of Virginia and Carolinas. The second objective (2) was to identify drought tolerance related genomic regions using three mapping populations. Phenotyping and genotyping of three diverse recombinant inbred line (RIL) populations, derived from crossing lines N05006 x N04074FCT (Pop-1), lines N05006 x Phillips, an old virginia-type cultivar (Pop-2), and lines N08086olJCT x PI 585005 (Pop-3) were done to find quantitative trait loci (QTL) for drought related traits. These population were phenotyped for the Normalized Difference Vegetation Index (NDVI), Canopy Temperature Depression (CTD), SPAD-meter relative chlorophyll content of the leaves (SPAD) and wilting for QTL mapping. These surrogate traits are related to trait of interest for drought tolerance. NDVI is effective in predicting biomass and yield. Similarly, CTD is associated with transpiration efficiency and carbon dioxide assimilation. Mapping identified 27 minor QTL on eight chromosomes for all physiological characteristics, i.e NDVI, CTD, SPAD and wilting with logarithmic of odds values range from 2.5 to 38.5 and the phenotypic variance explained by these traits ranging from 1.04 to 11.46 %. There were 4 loci on chromosome 2 associated with NDVI in Pop-1 and Pop-3, explaining 1.8 to 10.38% of the phenotypic variation. These genomic regions may be important resources in peanut breeding programs to improve drought tolerance. Further research is needed to increase the marker density in order to fine map the identified QTL and validate markers linked with these regions.

Doctor of Philosophy / Genetics Interdepartmental Program / Bikram S. Gill / Leaf rust, caused by fungus Puccinia triticina, is an important foliar disease of wheat worldwide. Breeding for race-nonspecific resistant cultivars is the best strategy to combat this disease. Aegilops tauschii, D genome donor of hexaploid wheat, has provided resistance to several pests and pathogens of wheat. To identify potentially new adult plant resistance (APR) genes, 371 geographically diverse Ae. tauschii accessions were evaluated in field with leaf rust (LR) composite culture of predominant races. Accessions from Afghanistan only displayed APR whereas both seedling resistance and APR were common in the Caspian Sea region. Seventeen accessions with high APR were selected for production of synthetic hexaploid wheat (SHW), using ‘TetraPrelude’ and/or ‘TetraThatcher’ as tetraploid parents. Six SHWs were produced and evaluated for APR to LR and resistance to tan spot at seedling stage. Genetic analysis and mapping of APR introgressed from accession TA2474 was investigated in recombinant inbred lines (RIL) population derived from cross between SHW, TA4161-L3 and spring wheat cultivar, ‘WL711’. Genotyping-by-sequencing approach was used to genotype the RILs. Maximum disease severity (MDS) for LR was significantly correlated among all experiments and APR to LR was highly heritable trait in this population. Nine genomic regions significantly associated with APR to LR were QLr.ksu-1AL, QLr.ksu-1BS, QLr.ksu-1BL.1, QLr.ksu-1BL.2, QLr.ksu-2DS, QLr.ksu-2DL, QLr.ksu-5AL, QLr.ksu-5DL and QLr.ksu-6BL. Association of QLr.ksu-1BL.1 with marker Xwmc44 indicated this locus could be slow-rusting APR gene, Lr46/Yr29. QTLs detected on 2DS, 2DL and 5DL were contributed by TA4161-L3 and are novel, along with QLr.ksu-5AL. Tan spot, caused by necrotrophic fungus, Pyrenophora tritici-repentis, has recently emerged as a damaging disease of wheat worldwide. To identify QTLs associated with resistance to Race 1 of P. tritici-repentis, F[subscript]2:3 population derived from cross between SHW, TA4161-L1 and winter wheat cultivar, ‘TAM105’ was used. Two major effect QTLs, QTs.ksu-1AS.1 and QTs.ksu-7AS were significantly associated with tan spot resistance and contributed by TA4161-L1. QTs.ksu-7AS is a novel QTL and explained 17% of the phenotypic variation. Novel QTLs for APR to LR and tan spot identified in SHWs add new variation for broadening the gene pool of wheat and providing resources for breeding of durable resistant cultivars.

<p>Cerebrovascular disease (stroke), especially ischemic stroke, is a major cause of death and neurological disability in adults. Because of its clinical heterogeneity, stroke is considered as a multi-factorial and polygenic disorder. Most current genetic studies of ischemic stroke focus on genetic susceptibility rather than factors determining stroke outcome. The genetic components of ischemic stroke outcome are difficult to study in humans due to environmental factors and medical intervention. Thus, we proposed to use a surgically induced, permanent, focal cerebral ischemic stroke mouse model to investigate genetic factors of ischemic stroke outcome measured by infarct volume. This model is the middle cerebral artery occlusion (MCAO) model. First, we screened infarct volumes across 32 inbred mouse strains. The infarct volume varies between strains, and this strongly suggests that infarct volume is genetically determined. To identify these genetic factors, we used genome-wide association study [Efficient Mixed-Model Association (EMMA) analysis] on infarct volume from 32 inbred mouse strains. Using the EMMA analysis, we identified 11 infarct volume-associated loci; however, most loci were mapped with missing alleles. This suggests that these loci might be false positives. Thus, we used specifically designed scripts of EMMA analysis with updated mouse SNP database to correct for potential false positives. The loci identified by the updated EMMA analyses will led us to the identification of genes involved in ischemic stroke outcome. </p><p> There are two major mechanisms were proposed to be determinants of infarct volume, the extent of native collateral circulation and neuroprotection. Using the infarct volume screening panel from 32 inbred strains, we observed that infarct volume is inversely correlated with the native collateral vessel number. However, among these inbred strains, we also observed several strains differ significantly in infarct volumes but harbor similar collateral numbers. In order to identify genetic factors determining infarct volume in a collateral-independent manner (neuroprotection), we used quantitative trait locus (QTL) mapping on mouse strains that exhibit the most difference in infarct volumes but the least difference in collateral numbers (C57BL/6J and C3H/HeJ). From the F2 B6 x C3H cross, we mapped 4 loci determining infarct volume (cerebral infarct volume QTL 4 to 7, Civq4 to Civq7). The Civq4 locus is the strongest locus (LOD 9.8) that contributes 21% of phenotypic variance in infarct volume. We also used a parallel F2 B6 x C3H cross to perform a QTL mapping on collateral vessel traits to further verify these collateral-independent loci. Among these 4 loci, the Civq4 and Civq7 loci appear to be truly collateral-independent. Based on strain-specific sequence variants and mRNA expression differences, we proposed Msr1 and Mtmr7 are the potential candidate genes of the Civq4 locus. Identification of the collateral-independent genetic factors will help to understand the genetic architecture, disease pathophysiology and potential therapeutic targets for of ischemic stroke</p> / Dissertation

L'arachide (Arachis hypogaea L.) est une plante allotétraploïde (2n = 4x = 40) issue d'un événement récent d'hybridation entre deux espèces sauvages suivi d'un doublement spontané des chromosomes. La variabilité génétique existant dans le compartiment cultivé est faible. Les espèces diploïdes sauvages apparentées à l'arachide cultivée représentent un important réservoir d'allèles nouveaux utilisables pour élargir le pool génique du cultigène. L'objectif général de cette thèse est d'augmenter les marges de progression en amélioration de l'arachide par recours aux ressources génétiques sauvages. Nous avons développé une population BC1F1 à partir du croisement entre Fleur11 et un amphidiploïde synthétique (A. ipaensis x A. duranensis)x4 qui associe les génomes des plus probables progéniteurs sauvages de l'espèce cultivée. Nous avons d'abord construit une carte génétique comprenant 298 loci positionnés sur 21 groupes de liaison avec une taille totale de 1843.7 cM. Nous avons ensuite conduit une analyse AB-QTL pour plusieurs caractères impliqués dans la productivité, l'adaptation et la domestication de l'arachide. Au total, nous avons cartographié 95 QTL. Pour plusieurs QTL, les effets positifs sont associés aux allèles provenant des espèces sauvages. Nous avons aussi identifié trois régions du génome qui portent des empreintes de la domestication. Nous avons enfin développé une population de 122 lignées d'introgression à l'aide d'une stratégie de sélection assistée par marqueurs. L'ensemble des groupes de liaison sont couverts avec des frag ments chevauchants, issus des donneurs sauvages, d'une taille moyenne de 39.2 cM et chaque lignée comprend en moyenne 1.2 fragments. Nous avons par ailleurs discuté l'utilisation de cette collection de lignées d'introgression pour des applications de sélection et de recherche. Nos résultats ouvrent de nouvelles perspectives pour l'amélioration de l'arachide par croisement avec les espèces sauvages apparentées. / Peanut (Arachis hypogaea L.) is considered to be an allotetraploid (2n = 4x = 40) originated from a single hybridization event between two wild diploids followed by spontaneous chromosomes duplication. Cultivated peanut harbours a limited genetic diversity. Peanut wild relatives represent an important source of novel alleles that could be used to broaden the gene pool of the cultigen. The general objective of this work is to enhance the rate of progress in peanut breeding using wild species resources.We developed a BC1F1 population from the cross between Fleur11 and a synthetic amphidiploid (A. ipaensis x A. duranensis)x4 that combines the genomes of the most probable wild progenitors of the cultigen. We first developed a SSR based genetic map of 298 loci on 21 linkage groups, spanning a total map distance of 1843.7 cM. We then conducted a detailed AB-QTL analysis for several traits involved in peanut productivity and adaptation as well as in the domestication syndrome. We mapped a total of 95 QTLs. About half of the QTL positive effects were associated with alleles of the wild parents and several QTLs involved in yield components were specific to the water-limited treatment. We identified three genomic regions which bear footprints of domestication. We finally developed, through a marker assisted backcross strategy, an exotic introgression library of 122 lines. This population, which has in average 1.2 fragments per line, allows covering all linkage groups with overlappi ng wild donor fragments of in average 39.2 cM length. The utilization of the exotic introgression lines library for breeding and research is discussed. Our findings open new avenues for peanut improvement using wild relatives.

Organisms are often adapted to their local environment, but the role of early life stages in adaptive differentiation among populations remains poorly known. The aim of my thesis was to investigate the contribution of early life stages to the magnitude and genetic basis of local adaptation, and to identify the underlying adaptive traits. For this, I used two natural populations of the annual plant Arabidopsis thaliana from Italy and Sweden, and a Recombinant Inbred Line (RIL) population derived from a cross between these populations. By combining greenhouse and field experiments, Quantitative Trait Loci (QTL) mapping, and path analysis, I examined (1) the genetic basis of seed dormancy, (2) the contribution of differential seedling establishment to local adaptation, (3) among-year variation in selection during seedling establishment, (4) direct and indirect effects of seed dormancy and timing of germination on fitness, and (5) the adaptive value of the seed bank. I found that both the level and the genetic basis of seed dormancy were affected by the maternal environment. One major-effect QTL was identified in all maternal environments, which overlaps with the dormancy gene DELAY OF GERMINATION 1 (DOG1). Selection through seedling establishment success contributed strongly to local adaptation and genetic tradeoffs, and varied among years. Variation in seedling establishment and overall fitness among RILs could be explained by genetically based differences in seed dormancy and timing of germination. Seed dormancy affected fitness throughout the life cycle, by affecting the proportion of germinated seeds, and indirectly via effects on timing of germination, plant size and flowering time. My results suggest that a considerable portion of A. thaliana seeds enter the seed bank. I found genetic differences in dormancy cycling behaviour between the two populations, which could contribute to local adaptation. The value of a seed bank should be higher at the Swedish study site than at the Italian study site due to lower rate of seed mortality in the soil. Overall, the results of this thesis demonstrate that early life stages contribute strongly to both the magnitude and the genetics of local adaptation.