This study was conducted to identify and map QTLs for yield components and heat tolerance of wheat in response to two kinds of heat treatment (short term-and long term-heat treatment) during seed formation in a set of 62 RILs derived from a cross of ‘7C’ (heat resistant variety) and ‘Seri M82’ (heat susceptible variety) in environmentally controlled growth rooms and field. Phenotypic variations of yield components (kernel number, kernel weight, spike number and grain filling duration) were evaluated as indicators of heat tolerance / susceptibility. Most of the phenotypic variations of yield components exhibited a normally distributed pattern in response to heat stress treatments. This suggests that the yield component responses to high temperature stress are likely quantitatively inherited. A transgressive segregation pattern compared to the two parents was observed in several yield traits. This suggests that genetic variation from optimal recombination from the two parents have occurred in the progeny population. The Pearson correlation coefficients revealed significant correlations between yield components. This suggests the probability of co-segregation of genes controlling each yield components. The ANOVA also revealed a significant genotype x environment effect on individual yield components in response to reproductive stage high temperature stress. The heritability of the individual yield components was low (0.42 to 25%, 0.1~ 2% for heat tolerance). One hundred two polymorphic SSRs markers among 323 SSRs markers tested were used to construct a linkage coverage and average interval distance of 1860.2 cM and 18.2 cM/marker, respectively. Eighty-one QTLs for yield components and 68 QTLs for heat tolerance were detected with high LOD values (2.50~8.35 for yield components, 2.51~ 9.37 for heat tolerance) and that explained significant phenotypic variations (7~40% for individual QTL for yield components, 2~40 % for individual heat tolerance QTLs) from seven individual environments and the four individual heat stress environments, respectively. Specifically the regions between wmc48 and wmc89, and between wmc622 and wmc332 on the chromosome 4A and 6A, respectively possessed QTLs for both yield components and heat tolerance from various environments.

As compared with other cereal grains, Sorghum bicolor shows very low levels ofprotein digestibility when exposed to proteolytic enzymes. Protein digestibility furtherdecreases when sorghum is cooked. It is speculated that low digestibility is the result ofextensive disulfide crosslinking in the - and -kafirins (storage proteins) surroundingthe endosperm protein bodies. The degree of crosslinking increases as sorghum iscooked, causing the highly digestible -kafirins found at the interior of protein bodies tobe locked within a tightly bound capsule, inaccessible to digestive enzymes. In thisresearch project, two major QTLs were found to be associated with proteindigestibility—one QTL unfavorably affecting digestibility and one QTL favorablyaffecting digestibility. By identifying the QTLs and the linked markers corresponding tothe highly digestible trait, breeders will be able to use marker-assisted selection toquickly and accurately identify highly digestible lines to advance in a breeding program.

As compared with other cereal grains, Sorghum bicolor shows very low levels ofprotein digestibility when exposed to proteolytic enzymes. Protein digestibility furtherdecreases when sorghum is cooked. It is speculated that low digestibility is the result ofextensive disulfide crosslinking in the - and -kafirins (storage proteins) surroundingthe endosperm protein bodies. The degree of crosslinking increases as sorghum iscooked, causing the highly digestible -kafirins found at the interior of protein bodies tobe locked within a tightly bound capsule, inaccessible to digestive enzymes. In thisresearch project, two major QTLs were found to be associated with proteindigestibility—one QTL unfavorably affecting digestibility and one QTL favorablyaffecting digestibility. By identifying the QTLs and the linked markers corresponding tothe highly digestible trait, breeders will be able to use marker-assisted selection toquickly and accurately identify highly digestible lines to advance in a breeding program.

Maize (Zea mays L), the third largest crop produced in the world is susceptible to pre and post contamination of aflatoxin. Aflatoxin is a secondary carcinogenic metabolite produced by Aspergillus flavus, an opportunistic fungus of maize that causes ear-rot and subsequent production of aflatoxin. Host plant resistance is one of the promising long term solution to combat aflatoxin contamination in maize grains. Resistance to aflatoxin is heritable and resistant germplasms that can be used to reduce aflatoxin resistance has been developed. Creating quantitative trait loci (QTL) mapping populations to identify regions within maize chromosomes contributing to aflatoxin resistance help identify markers that can be used for successful introgression of the QTL into desirable backgrounds. To further confirm and validate QTL detected in a mapping population, creating near isogenic lines (NILs) with identified QTL is important. Furthermore, stacking of validated QTL through recurrent selection (RS) will help to improve aflatoxin resistance in maize. In this study, an F2:3 mapping population was derived from crossing CML69, an aflatoxin resistant inbred genotype unrelated to other genotypes used in previous mapping population and Va35, an aflatoxin susceptible inbred germplasm adapted to southeastern U.S. The F2:3 families were phenotyped for aflatoxin at Mississippi State, MS and Lubbock TX in 2016 and 2017 and genotyped using 1331 polymorphic markers. Composite interval mapping (CIM) identified 16 significant QTL in all four environments on chromosomes 1, 2, 3, 4, 5, and 7, and these QTL explains between 4% - 18% of the phenotypic variation observed in the population. Fourteen of the 15 QTL co-locate with previously identified QTL, suggesting that they will be stable in different genetic backgrounds and environments, and one novel QTL will provide additional resistance. Sixteen single nucleotide polymorphism (SNP) markers linked to QTL identified in a previous genome wide association mapping (GWAS) study were also used to create Near Isogenic Lines (NILs) to validate their effect on aflatoxin resistance. Finally, recurrent selection (RS) populations were created to increase the frequency of aflatoxin-resistant alleles from an 8-way cross derived from 8 aflatoxin resistant inbred maize lines.

La méthode traditionnelle utilisée pour produire des vins mousseux (comme cava et champagne) est caractérisée par une seconde fermentation qui a lieu à l'intérieur de la bouteille. Cette deuxième fermentation présente des caractéristiques très spécifiques telles qu’une teneur élevée en éthanol, la pression croissante de CO2, une basse température et une faible disponibilité des nutriments. Dans cette thèse, on a tout d'abord analysé l'effet des facteurs environnementaux sur la cinétique de fermentation de Saccharomyces cerevisiae au cours de la seconde fermentation. Deuxièmement, on a analysé l'effet de pratiques communes, telles que l'ajout de nutriments au vin de base, sur la composition finale du vin mousseux. Enfin, nous avons cherché à identifier la base génétique de la variabilité phénotypique observée pendant la seconde fermentation en utilisant une approche Quantitative Trait Locus (QTL). Les résultats obtenus ont permis de déterminer la température, le vin de base utilisé, la souche de levure et la source d'azote utilisée dans l'acclimatation de levure comme les facteurs qui ont la plus grande incidence dans la cinétique de la seconde fermentation. Deuxièmement, par rapport à la composition finale du vin mousseux, nous avons trouvé que l'addition d'azote dans le vin base favorise la libération des acides aminés. Bien que l'ajout de levure sèche inactive favorise la libération des polysaccharides et les propriétés moussantes du vin mousseux. Enfin, on a pu identifier quatre gènes dont la variation allélique explique la variation phénotypique observée parmi les souches. / The traditional method used to produce sparkling wines (such as cava and champagne) is characterized by a second fermentation that takes place inside the bottle. This second fermentation has very specific characteristics such as a high ethanol content, increasing CO2 pressure, low temperature and low nutrient availability. In this thesis, we have firstly analyzed the effect of environmental factors on fermentation kinetics of Saccharomyces cerevisiae during the second fermentation, by monitoring the second fermentation development using aphrometers. Secondly, we analyzed what is the effect of common practices such as adding nutrients to the base wine on the final composition of the sparkling wine by HPLC analysis of content of amino acids and polysaccharides and its foaming capacity (mosalux) of the sparkling wine. Finally we aimed to identify the genetic basis of the second fermentation using Quantitative Trait Locus (QTL) mapping and validation approach. The results obtained enabled us to identify the temperature, the base wine used, the yeast strain and source of nitrogen used in the acclimatization of yeast as the factors that have the highest impact in the second fermentation kinetics. Secondly, with respect to the final composition of sparkling wine, we have found that the addition of nitrogen to the wine base favors the release of amino acids. While the addition of inactive dry yeast, promotes the release of polysaccharides and favors the foaming properties of the sparkling wine. Finally, could identify four genes whose allelic variation explains the phenotypic variation observed among strains.

This work focuses on the evaluation of a bagging EB method in terms of its ability to select a subset of QTL-related markers for accurate EBV prediction. Experiments were performed on several simulated and real datasets consisting of SNP genotypes and phenotypes. The simulated datasets modeled different dominance levels and different levels of background noises. Our results show that the bagging EB method is able to detect most of the simulated QTL, even with large background noises. The average recall of QTL detection was $0.71$. When using the markers detected by the bagging EB method to predict EBVs, the prediction accuracy improved dramatically on the simulation datasets compared to using the entire set of markers. However, the prediction accuracy did not improve much when doing the same experiments on the two real datasets. The best accuracy of EBV prediction we achieved for the dairy dataset is 0.57 and the best accuracy for the beef dataset is 0.73.

Waves of bovine genomic data have been produced as a result of the bovine genome sequencing projects. In addition to the massive amounts of genomic sequence, significant annotation including single nucleotide polymorphisms, sequence tagged sites and haplotype blocks have been produced by the Bovine HapMap Project. Furthermore, many agriculturally significant traits in cattle such as milk yield and carcass weight are measured on a quantitative scale and have been genetically mapped as quantitative trait loci (QTL). QTL data can be used to generate another form of bovine annotation linking phenotype to genotype. However, it is impossible for humans to be able to analyze genomic scale data without computer based tools. Bioinformatic tools have been shown to greatly increase productivity and improve efficiency when dealing with large data sets. My dissertation presents an integrated, extensible database that houses SNPs, STSs, haplotypes, and QTL. The database is presented to researchers through a restructured, object oriented Bovine QTL Viewer that displays multiple levels of bovine annotation synergistically. Evaluation of use of the viewer was performed using a survey based approach and measured quantitatively. In addition, the QTL data from the database was used to analyze the frequency of gene ontology (GO) annotations within QTL regions. QTL regions were divided into 8 trait based groups. GO terms were counted within each category of QTL and in non- QTL regions of the genome. Top level GO term frequencies were generated from the counts and these frequencies were compared between QTL and non-QTL portions of the genome. Furthermore, specific sets of GO terms believed to be related to QTL categories were also used to determine if QTL regions were enriched for genes annotated with such GO terms. As a result, we determined that gene density varied significantly across QTL regions and that many QTL categories showed GO term frequency differences that could be related to the trait’s biology. Furthermore, our selected GO term sets were shown to be significantly enriched in some QTL categories.

I examined the genetic architecture and evolutionary significance of the considerable morphological and life history variation in Arctic charr (Salvelinus alpinus) from natural and cultured populations in Iceland. I found that the sex determining locus in Icelandic Arctic charr is located on a different linkage group relative to the majority of the Atlantic lineage Arctic charr, including populations from the Fraser River in Labrador Canada and Swedish and Norwegian strains. In addition, there may be a possible conservation of a sex linkage arrangement in Icelandic Arctic charr and Atlantic salmon. These observations suggest that the differentiation of the sex determination chromosome in salmonids is still in the early stages. I then tested hypotheses regarding the genetic architecture of wild and cultured populations of Icelandic Arctic charr identifying the number and effect size of quantitative trait loci (QTL) and their conservation within and across salmonid species. QTL with genome-wide significance for body size, condition factor and age of maturation in cultured fish from the Icelandic breeding program were detected on multiple linkage groups. Comparisons with a North American cultured strain of Arctic charr and North American populations of Atlantic salmon and rainbow trout revealed some conservation in QTL. Additionally, I compared the genetic architecture of fork lengths in juvenile wild and cultured populations and found moderate conservation of genomic regions among Icelandic populations. I compared linear measurements taken on the cleared and stained heads of benthic and limnetic morphs from two lake populations expressing varying degrees of divergence. I found high genetic variation for craniofacial morphology in all morphs with no significant difference of plasticity levels between them. However, stronger family effects and weaker morph effects in the less derived lake population suggest higher genetic variation for craniofacial traits compared to the more derived system. Overall QTL (suggestive and significant) number and effect size was similar but the morphs hypothesized to be more derived within each lake had about half the number of significant QTL (p<0.01) compared to their sympatric counterpart, suggesting fixation of alleles through canalization.

Alpha (α-) tocopherol is the most active component of vitamin E and is a highly desirable and heritable trait in soybean seed. The objectives of this study were to identify and validate regions of the soybean genome that correspond to tocopherol accumulation in seed. A mapping population (07-225) of 190 RILs was derived from a cross between Keszthelyi A.S. x 2355RR which identified that α-tocopherol expression in soybean seed was strongly associated with the Satt381 – Satt196 region of chromosome 9 (LG K) in lines with Keszthelyi A.S. in their pedigree. Two other F4:5 validation populations derived from the crosses Keszthelyi A.S. x PRO 2995R (population 07-223; 94 RILs) and Keszthelyi A.S. x 2355RR (population 07-226; 45 RILs) were used to validate QTL in literature. The markers Satt079 (Ch 6; LG C2), Satt440 (Ch 20; LG I) and Satt381 (Ch 9; LG K) were significantly associated with α-tocopherol expression in these populations. / OMAFRA Food Program, Grain Farmers of Ontario, CanAdvance, Hendrick Seeds

No description available.