Application and Development of Genomic Techniques in Forest Trees

This dissertation has three separate research projects. The first project proposes a mechanism that would explain why hybridization is so prevalent in oaks (Quercus) which hybridize freely within the same botanical section. Hybrid zones do not usually form, and the hybrids are fertile and capable of backcrossing. Due to its prevalence and lack of species barriers closing between species, we propose that hybridization must yield some fitness advantage or benefit to the parent species. We test this by subjecting a population of progeny seedlings from three species (Quercus bicolor, Q. macrocarpa, and Q. robur) and two hybrids (Q. x schuettei (Q. bicolor x Q. macrocarpa) and Q. x warei (Q. robur x Q. macrocarpa)) to a severe drought test in a greenhouse. Our data shows that while under severe drought conditions the hybrids survive better than both or one of its parent species. Under normal conditions they hybrids are as fit as or less fit than the parent species. We use these data to inform a basic drought model testing whether the results we observe could create an episodic hybrid advantage when following historical drought cycles in June in Missouri (1895-2018). We term this the Hybrid Episodic Advantage Theory (HEAT).

The next data chapter is aimed at developing a panel of ancestry informative markers to differentiate the ancestry of the many cultivated chestnuts in production today. Most chestnut species are able form fertile hybrids with each other. American Chestnut (Castanea dentata) populations were decimated in the early 20^th century by chestnut blight (Cryphonectria parasitica (Murrill) Barr) which rendered the species ecologically extinct. Various unsuccessful breeding efforts have taken place attempting to create a blight-resistant mostly-American chestnut and reintroduce it back into its native environment. However, years of uncontrolled hybridization in cross-species orchards and chestnut-blight resistance breeding programs have resulted in a myriad of cultivars with unknown genetic background. We aim to solve this issue by developing a set of genetic markers that can be used to cheaply screen the recent ancestry of any chestnut tree. We generated these genetic markers by harvesting EST-SSR sequences from publicly available transcriptomic data. We worked with ArborBiosciences to create a target-capture platform to sequence hundreds of EST-SSR markers in parallel.

The final data chapter revolves around adding EST-SSR markers to a previously constructed SNP-based genetic map for a mapping population of 282 full-sibs in black walnut (Juglans nigra). These markers were developed using the target-capture technology used in Chapter 3 for 61 of the 282 individuals. In total, 54 markers were added to the linkage map. More are expected to map as more individuals are sequenced later this year. Juglans species exhibit a high degree of collinearity, so a well-constructed map in Juglans nigra will be useful in Juglans regia and other species.Overall, these projects will add significant resources to their respective communities and can provide long term benefits beyond the scope of this dissertation.