Atlantic salmon is a species of both scientific and economic importance, and Atlantic salmon farming is a highly profitable industry worldwide. One of the biggest challenges being faced by farms, which affects production efficiency and results in severe economic loss, is disease. In livestock production, one of the approaches taken to limit the impact of disease outbreaks is to selectively breed for improved resistance within farmed populations. Although traditional family-based resistance breeding programs have shown improvements in resistance to a variety of bacterial, viral and parasitic diseases on Atlantic salmon farms, response to selection can be slow. One way of increasing selection efficiency is through the incorporation of genetic markers into breeding programs, for marker-assisted or genomic selection. However, genomic resources for cultured aquatic species are sparse, and the generation of new and denser resources for use in selective breeding programs would be advantageous. The main focus of this thesis is the development of genomic resources in Atlantic salmon and the application of those resources to gain a better understanding of the salmon genome, particularly in the genetic basis of host resistance to infectious diseases. The first aim of this thesis was to develop improved genomic resources for Atlantic salmon, and to characterise the Atlantic salmon genome via construction and analysis of a SNP linkage map derived from RAD-Sequencing (RAD-Seq). Approximately 6,500 SNPs were assigned to 29 linkage groups, and ~1,800 male-segregating, and ~1,400 female-segregating SNPs were ordered and positioned. Overall map lengths and recombination ratios were relatively consistent between the sexes and across the linkage groups (~1:1.5, male:female). However, a substantial difference in the degree of marker clustering was seen between males and females, which is reflective of the difference in the positions of chiasmata between the two sexes. Using this map, ~4,000 Atlantic salmon reference genome contigs were assigned to a linkage group, and 112 contigs were assigned to multiple linkage groups, highlighting regions of homeology (large sections of duplicated chromosomal regions) within the salmon genome. Alignment of SNP-flanking sequences to the stickleback and rainbow trout genomes identified putative gene-associated SNPs and cross-species chromosomal orthologies, and provided evidence in support of the salmonid-specific genome duplication. In addition, based on this and other publically available RAD-Seq datasets, the utility of RAD-Seq-derived data from different species and laboratories for population genetics analyses was tested. Short RAD-Seq contigs in Atlantic salmon and nine other teleost fish were used to identify cross-species orthologous genomic relationships. Several thousands of orthologous RAD loci were identified across the species, with the number of RAD loci decreasing with evolutionary distance, as expected. Previously published broad-level relationships between orthologous chromosomes were confirmed. The identified cross-species orthologous RAD loci were used to estimate evolutionary relationships between the ten teleost fish species. Previously published relationships were recovered, suggesting that RAD-Seq data derived from different laboratories is useful for this purpose. The second aim was to characterise the genetic architecture of resistance to two viral diseases affecting Atlantic salmon production on farms: pancreas disease (PD), and infectious pancreatic necrosis (IPN). Using data and samples collected from a large population of salmon fry challenged with PD, a high heritability for resistance was estimated (h2 ~0.5), and four QTL were identified, on chromosomes 3, 4, 7 and 23. The QTL explaining the highest within-family variation for resistance was located on chromosome 3. This QTL has been confirmed in a population of post-smolts by an independent research group, highlighting the potential for its incorporation into breeding programs to improve PD resistance. For IPN, the major resistance QTL had previously been mapped to linkage group 21. However, the mutation(s) underlying this QTL effect and the consequences of these mutation(s) on the affected genes and relevant biological resistance mechanisms are unknown. To generate a list of candidate genes within the vicinity of the IPN QTL, QTL-linked DNA sequences were aligned to four model fish genomes. This identified two QTL-orthologous regions in each of the species, and gene order within these regions was highly conserved across species. Analysis of gene expression patterns between IPN resistant and susceptible salmon in a viral challenge experiment revealed that the five most significantly differentially-expressed genes mapped to the QTL-orthologous region on linkage group II of stickleback. Pathway enrichment analysis across all differentially-expressed genes suggests that biological pathways influencing viral infection stress response/entry/replication, cellular energy production and apoptosis may be involved in resistance during the initial stages of IPN virus (IPNV) infection. These results have provided the basis for further study of the putative involvement of these candidate genes and pathways in genetic resistance to IPNV. In summary, the results and resources presented in this thesis extend our current understanding of the salmon genome and the genetic basis of resistance to two viral diseases, and provide resources with the potential to be used in Atlantic salmon selective breeding programs to tackle disease outbreaks.

Modern healthcare research demands collaboration across disciplines to build preventive measures and innovate predictive capabilities for curing diseases. Along with the emergence of cutting-edge computational and statistical methodologies, data generation and analysis has become cheaper in the last ten years. However, the complexity of big data due to its variety, volume, and velocity creates new challenges for biologists, physicians, bioinformaticians, statisticians, and computer scientists. Combining data from complex multiple profiles is useful to better understand cellular functions and pathways that regulates cell function to provide insights that could not have been obtained using the individual profiles alone. However, current normalization and artifact correction methods are platform and data type specific, and may require both the training and test sets for any application (e.g. biomarker development). This often leads to over-fitting and reduces the reproducibility of genomic findings across studies. In addition, many bias correction and integration approaches require renormalization or reanalysis if additional samples are later introduced. The motivation behind this research was to develop and evaluate strategies for addressing data integration issues across data types and profiling platforms, which should improve healthcare-informatics research and its application in personalized medicine. We have demonstrated a comprehensive and coordinated framework for data standardization across tissue types and profiling platforms. This allows easy integration of data from multiple data generating consortiums. The main goal of this research was to identify regions of genetic-epigenetic co-ordination that are independent of tissue type and consistent across epigenomics profiling data platforms. We developed multi-‘omic’ therapeutic biomarkers for epigenetic drug efficacy by combining our biomarker regions with drug perturbation data generated in our previous studies. We used an adaptive Bayesian factor analysis approach to develop biomarkers for multiple HDACs simultaneously, allowing for predictions of comparative efficacy between the drugs. We showed that this approach leads to different predictions across breast cancer subtypes compared to profiling the drugs separately. We extended this approach on patient samples from multiple public data resources containing epigenetic profiling data from cancer and normal tissues (The Cancer Genome Atlas, TCGA; NIH Roadmap epigenomics data).

The human bronchial epithelium is composed of multiple, discrete cell types that cooperate to perform mucociliary clearance. While previous studies have shown that cigarette smoke can alter bronchial epithelial gene expression, the underlying effects of this exposure on specific cell types are not well understood. In this thesis, single-cell RNA sequencing was used to profile bronchial epithelial cells from six current smokers and six never smokers. Thirteen cell clusters were identified that were defined by expression of unique combinations of nineteen distinct gene sets. This clustering revealed that smoke exposure induced expression of a toxin metabolism program that specifically associated with ciliated cells. Extensive airway remodeling was also observed, in which smoking was associated with loss of club cells as well as goblet cell expansion and hyperplasia. Additionally, a previously uncharacterized CEACAM5+ KRT8+ epithelial subpopulation was identified in the airways of smokers. While it has been shown that most smoking-associated gene expression alterations can be reversed upon smoking cessation, a subset of these alterations persists in former smokers. The basal layer of the bronchial epithelium is comprised of a multipotent progenitor subpopulation. When abnormalities persist in the bronchial epithelium despite normal tissue turnover, the source of these abnormalities may be traced to this progenitor population and its program of differentiation. Therefore, basal cells were procured from three current smokers and three never smokers, differentiated in vitro, and profiled by RNA sequencing at eight time points spanning the differentiation procedure. Twenty-seven unique sets of co-expressed genes associated with differentiation were identified and functionally characterized, a subset of which were abnormally expressed in smoker cells. Robust expression of genes involved with the unfolded protein response was specifically detected in smoker basal cells. Additionally, a smoking-associated delay in the onset of expression of genes involved with ciliogenesis was observed. These data therefore indicate that smoking has long-term consequences on the differentiated state of the airway epithelium. Collectively, the observations outlined in this thesis demonstrate that smoking drives a complex landscape of alterations that affects the function and composition of the human bronchial epithelium. / 2020-10-24T00:00:00Z

Les bactéries du phylum Bacteroidetes, classe Bacteroidia, sont parmi les plus importantes dans microbiotes gastrointestinaux des humains et d'autres mammifères. La bouche, entrée du tube digestif, est un environnement avec des sites anatomiques variés, auxquels s'associent des microbiotes de composition différente. L'union de la gencive et des dents, le sillon gingivo-dentaire ou sulcus, est un site de dépôt d'un biofilm complexe appelé plaque dentaire. Une bactérie de ce phylum, Porphyromonas gingivalis, est capable de perturber le système immunitaire humain et de produire un déséquilibre du biofilm oral également nommée dysbiose. Ceci déclenche la formation de la poche parodontale, un creusement pathologique du sulcus, et l'apparition de la parodontite. D’autres espèces du genre Porphyromonas sont également associées à la parodontite notamment chez les canidés. Les populations de P. gingivalis sont panmictiques et la plasticité de leurs génomes importante. La bioinformatique peut aider à identifier les causes de la mosaïcité des génomes de cette bactérie, à étudier les facteurs de virulence au niveau du genre bactérien pour expliquer l'existence d'espèces pathogènes et d'autres commensales et à décrire la dysbiose liée à la parodontite. La génomique comparative de P. gingivalis a démontré une corrélation entre le nombre de contigs dans les génomes draft de cette espèce et les répétitions génomiques, notamment des séquences d'insertion. Nous avons re-séquencé, re-assemblé et re-annoté trois souches de référence de cette bactérie qui avaient des génomes complets, en utilisant un séquençage en long-read. Nous avons mis en évidence des erreurs d'assemblage sur les trois génomes publiés, que nous avons corrigé. Une étude du pangénome de ces trois souches montre un génome core important. La plasticité de l'espèce serait donc plus dans l'organisation du génome que dans les différentes capacités de codage. Une sous partie du génome core, dont les gènes ont un pourcentage d'identité nucléotidique plus faible que la plupart (génome core variant) est intéressante pour expliquer les différences phénotypiques de ces bactéries. Nous avons étudié la répartition d'un facteur de virulence, les fimbriae, structures d'adhésion, au sein du genre Porphyromonas et lié les loci à la phylogénie et au caractère pathogène des espèces. Finalement, une description de la dysbiose qui a lieu lors d'une parodontite est faite par une analyse du microbiote de patients atteints de parodontite et d'individus sains. Les genres prépondérants lors des deux états sont mis en évidence. Au cours de ces travaux, nous montrons l'importance de la biocuration et sa valeur ajoutée dans les travaux de génomique et bioinformatique en général. Seulement en faisant ce travail lent et lourd de biocuration, les réponses apportées aux questions biologiques seront pertinentes. / Bacteria of Bacteroidetes phylum, Bacteroidia class, are amongst the more important in gastrointestimal microbiota, either human or from other mammals. The mouth, digestive tube entry, is an environment with varied anatomic sites, each having a particular microbiota with different composition. The union between gingiva and teeth, the gingival sulcus, is a site for biofilm (dental plaque) formation and accumulation. Porphyromonas gingivalis, a bacterium from this phylum, can modulate the inmune system and produce an oral biofilm desequilibrium called dysbiosis. This triggers the formation of a periodontal pocket, a pathological deepening of the gingival sulcus, and the emergence of periodontitis. Other Porphyromonas species are also associated to periodontitis, mainly in canids. P. gingivalis populations are panmictic and their genomes are highly plastic. Bioinformatics can help to identify the causes of this genomic mosaicity, to study Porphyromonas virulence factors in order to explain why some species are pathogens and other are commensal, and to describe the dysbiosis linked to periodontitis. P. gingivalis comparative genomics showed a correlation between the number of contigs in draft genomes and genomic repeats, mainly insertion sequences. We resequenced, reassembled and reannotated three reference strains of this bacterium that already had complete published genomes, using long-read sequencing. We showed that misassemblies were present in the three published genomes, and we corrected them. A pangenome study of the three strains showed that the core genome is preponderant. The species plasticity might be related more to the genome organization than to different coding capacities. A subpart of th core genome, with genes having a nucleotidic identity percentage lower than the majority (variable core genome), is interesting for explaining the phenotypic differences of bacteria. We analysed the repertoire of a virulence factor, fimbriae, adhesion structures, in the Porphyromonas genus to link the loci to phylogeny and pathogenicity of its species. Finally, we described the dysbiosis occuring with periodontitis, analysing gingival microbiota of patients having the illness and healthy individuals. Preponderant genera in both states are highlighted. With this work, we demonstrate the importance of biocuration and its added value for genomic and bioinformatic studies in general. Only with this slow and arduous work, the answers to biological questions will be relevant.

The ancient continent of Sahul, encompassing Australia, New Guinea and Tasmania, contains some of the earliest archaeological evidence for humans outside of Africa, dating back to at least 50 thousand years ago (kya). New Guinea was also one of the sites were humans developed agriculture in the last 10 thousand years. Despite the importance of this part of the world to the history of humanity outside Africa, little is known about the population history of the people living here. In this thesis I present population-genetic studies using whole-genome sequencing and genotype array datasets from more than 500 indigenous individuals from Australia and New Guinea, as well as initial work on large-scale sequencing of other, worldwide, human populations in the Human Genome Diversity Project panel. Other than recent admixture after European colonization of Australia, and Southeast Asian ad- mixture in the lowlands of New Guinea in the last few millennia, the populations of Sahul appear to have been genetically independent from the rest of the world since their divergence ∼50 kya. There is no evidence for South Asian gene flow to Australia, as previously suggested, and the highlands of Papua New Guinea (PNG) have remained unaffected by non-New Guinean gene flow until the present day. Despite Sahul being a single connected landmass until ∼8 kya, different groups across Australia are nearly equally related to Papuans, and vice versa, and the two appear to have separated genetically already ∼30 kya. In PNG, all highlanders strikingly appear to form a clade relative to lowlanders, and population structure seems to have been reshaped, with major population size increases, on the same timescale as the spread of agriculture. However, present- day genetic differentiation between groups is much stronger in PNG than in other parts of the world that have also transitioned to agriculture, demonstrating that such a lifestyle change does not necessarily lead to genetic homogenization. The results presented here provide detailed insights into the population history of Sahul, and sug- gests that its history can serve as an independent source of evidence for understanding human evolutionary trajectories, including the relationships between genetics, lifestyle, languages and culture.

A fundamental challenge in biology is to understand the complex gene regulatory networks which control tissue development in the mammalian embryo, and maintain homoeostasis in the adult. The cell fate decisions underlying these processes are ultimately made at the level of individual cells. Recent experimental advances in biology allow researchers to obtain gene expression profiles at single-cell resolution over thousands of cells at once. These single-cell measurements provide snapshots of the states of the cells that make up a tissue, instead of the population-level averages provided by conventional high-throughput experiments. The aim of this PhD was to investigate the possibility of using this new high resolution data to reconstruct mechanistic computational models of gene regulatory networks. In this thesis I introduce the idea of viewing single-cell gene expression profiles as states of an asynchronous Boolean network, and frame model inference as the problem of reconstructing a Boolean network from its state space. I then give a scalable algorithm to solve this synthesis problem. In order to achieve scalability, this algorithm works in a modular way, treating different aspects of a graph data structure separately before encoding the search for logical rules as Boolean satisfiability problems to be dispatched to a SAT solver. Together with experimental collaborators, I applied this method to understanding the process of early blood development in the embryo, which is poorly understood due to the small number of cells present at this stage. The emergence of blood from Flk1+ mesoderm was studied by single cell expression analysis of 3934 cells at four sequential developmental time points. A mechanistic model recapitulating blood development was reconstructed from this data set, which was consistent with known biology and the bifurcation of blood and endothelium. Several model predictions were validated experimentally, demonstrating that HoxB4 and Sox17 directly regulate the haematopoietic factor Erg, and that Sox7 blocks primitive erythroid development. A general-purpose graphical tool was then developed based on this algorithm, which can be used by biological researchers as new single-cell data sets become available. This tool can deploy computations to the cloud in order to scale up larger high-throughput data sets. The results in this thesis demonstrate that single-cell analysis of a developing organ coupled with computational approaches can reveal the gene regulatory networks that underpin organogenesis. Rapid technological advances in our ability to perform single-cell profiling suggest that my tool will be applicable to other organ systems and may inform the development of improved cellular programming strategies.

Understanding how genetic variation propagate to differences in phenotypes in individuals is an ongoing challenge in genetics. Genome-wide association studies have allowed for the identification of many trait-associated genomic loci. However, they are limited in their inability to explain the altered cellular mechanism. Genetic variation can drive disease by altering a range of mechanisms, including signalling networks, TF binding, and protein folding. Understanding the impact of variants on such processes has key implications in therapeutics, drug development, and more. This thesis aims to utilise computational predictors to shed light on how cellular mechanisms are altered in the context of genetic variation and better understand how they drive both molecular and organism-level phenotypes. Many binding events in the cell are mediated by short stretches of sequence motifs. The ability to discover these underlying rules of binding could greatly aid our understanding of variant impact. Kinase–substrate phosphorylation is one of the most prominent post-translational modifications (PTMs) which is mediated by such motifs. We first describe a computational method which utilises interaction and phosphorylation data to predict sequence preferences of kinases. Our method was applied to 57% of human kinases capturing known well-characterised and novel kinase specificities. We experimentally validate four understudied kinases to show that predicted models closely resemble true specificities. We further demonstrate that this method can be applied to different organisms and can be used for other phospho-recognition domains. The described approach allows for an extended repertoire of sequence specificities to be generated, particularly in organisms for which little data is available. TF-DNA binding is another mechanism driven by sequence motifs, which is key for the tight regulation of gene expression and can be greatly altered by genetic variation. We have comprehensively benchmarked current methods used to predict non-coding variant effects on TF-DNA binding by employing over 20,000 compiled allele-specific ChIP-seq variants across 94 TFs. We show that machine learning-based approaches significantly outperform more rudimentary methods such as the position weight matrix. We further note that models for many TFs with distinct binding specificities were unable to accurately assess the impact of variants. For these TFs, we explore alternative mechanisms underlying TF-binding, such as methylation, co-operative binding, and DNA shape that drive poor performance. Our results demonstrate the complexity of predicting non-coding variant effects and the importance of incorporating alternative mechanisms into models. Finally, we describe a comprehensive effort to compile and benchmark state-of-the-art sequence and structure-based predictors of mutational consequences and predict the effect of coding and non-coding variants in the reference genomes of human, yeast, and E. coli. Predicted mechanisms include the impact on protein stability, interaction interfaces, and PTMs. These variant effects are provided through mutfunc, a fast and intuitive web tool by which users can interactively explore pre-computed mechanistic variant impact predictions. We validate computed predictions by analysing known pathogenic disease variants and provide mechanistic hypotheses for causal variants of unknown function. We further use our predictions to devise gene-level functionality scores in human and yeast individuals, which we then used to perform gene-phenotype associations and uncover novel gene-phenotype associations.

Advances in DNA synthesis technology have led to rapid growth in the field of synthetic biology, heralding a nascent era of synthetic genomics. Sc2.0 (Saccharomyces cerevisiae version 2.0) is an international consortium with the aim of designing and constructing a fully‐synthetic eukaryotic genome. Fundamental design changes to the synthetic genome include the removal of unstable tRNA genes and their intended collation onto a “tRNA neochromosome”, with the aim of producing a more robust and stable synthetic genome structure. To maintain viability of a synthetic yeast, the tRNA neochromosome is therefore considered an important if not essential aspect of this project. The application of engineering principles is synonymous with synthetic biology, regularly employing the recursive Design‐Build‐Test cycle to improve experimental approach. This doctoral study explores the design, construction and characterisation of a tRNA neochromosome in Saccharomyces cerevisiae. A series of design principles influenced by engineering concepts were used to rationalise the complexities of de novo chromosome engineering, maximise its stability and ensure function in vivo. A methodology based on in vivo homologous recombination was then developed and shown to reliably construct the neochromosome from its constituent parts. Experimental characterisation revealed that genetic elements function as expected, and that the parental strain can tolerate the sole presence of one each of three single‐copy, essential tRNA genes (SUP61, TRT2 and TRR4), although Northern blot revealed potential precursor accumulation of the SUP61 tRNA caused by the presence of a synthetic 5’ flanking sequence. Following the addition of synthetic telomere seed sequences, pulsed‐field gel electrophoresis (PFGE) and deep sequencing revealed complex structure variations in two independent strain backgrounds. Except for these structural variations, successful neochromosome construction demonstrated the applicability of the approaches used and the remarkable ability of the yeast model to support the presence of a 17th chromosome housing an additional 275 tRNA genes. The research in this thesis has for the first time described the design, construction and characterisation of a eukaryotic neochromosome de novo. It is hoped that the findings presented will further our understanding of tRNA biology and enhance the aims of the Sc2.0 project.

xi, 52 p. ; ill. (some col.) A print copy of this title is available through the UO Libraries under the call number: SCIENCE QH445.2 .B35 2008 / Activation of Heat shock proteins (Hsps) is critical to adaptation to low oxygen levels (hypoxia) and enduring the oxidative stress of reoxygenation. Hsps are known to be regulated by Heat shock factor (Hsf), but my results demonstrate an unexpected regulatory link between the oxygen sensing and heat shock pathways. Hsf transcription is upregulated during hypoxia due to direct binding by Hypoxia-inducible Factor-1 (HIF-1) to HIF-1 response elements in an Hsf intron. This increase in Hsf transcripts is necessary for full Hsp induction during hypoxia and reoxygenation. The HIF-1-dependent increase in Hsps has a functional impact, as reduced production of Hsps decreases viability of adult flies exposed to hypoxia and reoxygenation. Thus, HIF-1 control of Hsf transcriptional levels is a regulatory mechanism for sensitizing heat shock pathway activity in order to maximize production of protective Hsps. This cross-regulation represents a mechanism by which the low oxygen response pathway has assimilated complex new functions by regulating the heat shock pathway's key transcriptional activator. Beyond studying the regulation of specific genes. I have also developed a method to identify small, yet important, changes within entire genomes. Genetic variation is the foundation of phenotypic traits, as well as many disease states. Variation can be caused by inversions, insertions, deletions, duplications, or single nucleotide polymorphisms (SNPs) within a genome. However, identifying a genetic change that is the cause of a specific phenotype or disease has been a difficult and laborious task for researchers. I developed a technique to quickly and accurately map genetic changes due to natural phenotypic variation or produced by genetic screens. I utilized massively parallel, high-throughput sequencing and restriction site associated DNA (RAD) markers, which are short tags of DNA adjacent to the restriction sites. These RAD markers generate a genome-wide signature of fragments for any restriction enzyme. Taken together with the fact that the vast majority of organisms have SNPs that disrupt restriction site sequences, the differences in the restriction fragment profiles between individuals can be compared. In addition, by using bulk segregant analysis, RAD tags can be used as high-density genetic markers to identify a genetic region that corresponds to a trait of interest. This dissertation includes both previously published and unpublished co-authored materials. / Adviser: Eric Johnson

The advent of high-throughput DNA and RNA sequencing has made possible the assay of millions of nucleic acid molecules in parallel. This allows functional genomic elements to be identified from background in single-tube experiments. This dissertation discusses the development of two such functional screens as well as work implementing a third that was previously developed in my thesis laboratory. Restriction-Associated DNA sequencing (RAD-Seq) is a complexity reduction sequencing method that allows the same subset of genomic sequence to be read across multiple samples. Differences in sample collection and data analysis allow manifold applications of RAD-Seq. Here we use RAD-Seq to identify mutant genes responsible for altered phenotypes in Caenorhabditis elegans and to identify hyper-invasive alleles in trout population admixtures. Apart from acquiring genomic sequence data, massively-parallel sequencing can be used for counting applications that quantify activity across a large number of test molecules. This dissertation describes the development of a technique for simultaneously quantifying the activity of a restriction enzyme across all possible DNA substrates by linking digest of a sequenced genome to Illumina-sequencing in an unbiased fashion. Finally, a powerful approach to analyze transcriptional activation is described. This method quantifies output from millions of potential DNA transcriptional enhancers via RNA amplicon sequencing of covalently-linked randomer tags and is used in conjunction with RNA-Seq to provide a mechanistic view of hypoxic gene regulation in Drosophila. This dissertation includes previously published, co-authored material