Unravelling the genetic architecture of common diseases is a continuing challenge in human genetics. While genome-wide association studies (GWAS) have proven to be successful in identifying many new disease susceptibility loci, the extension of these studies beyond single-SNP methods of analysis has been limited. The incorporation of multi-locus methods of analysis may, however, increase the power of GWAS to detect genes of smaller effect size, as well as genes that interact with each other and the environment. This investigation carried out large-scale simulations of four multi-locus model selection techniques; namely forward and backward selection, Bayesian model averaging (BMA) and least angle regression with a lasso modification (lasso), in order to compare the type I error rates and power of each method. At a type I error rate of ~5%, lasso showed the highest power across varied effect sizes, disease frequencies and genetic models. Lasso penalized regression was then used to perform three different types of analysis on GWAS data. Firstly, lasso was applied to the Wellcome Trust Case Control Consortium (WTCCC) data and identified many of the WTCCC SNPs that had a moderate-strong association (p<10-5) type 2 diabetes (T2D), as well as some of the moderate WTCCC associations (p<10-4) that have since been replicated in a large-scale meta-analysis. Secondly, lasso was used to fine-map the 17q21 childhood asthma risk locus and identified putative secondary signals in the 17q21 region, that may further contribute to childhood asthma risk. Finally, lasso identified three potential interaction effects potentially contributing towards coronary artery disease (CAD) risk. While the validity of these findings hinges on their replication in follow-up studies, the results suggest that lasso may provide scientists with exciting new methods of dissecting, and ultimately understanding, the complex genetic framework underlying common human diseases.

Doctor of Philosophy / Department of Agronomy / Guihua Bai / Allan K. Fritz / Wheat pre-harvest sprouting (PHS), germination of physiologically matured grains in a wheat spike before harvesting, can cause significant reduction in grain yield and end-use quality. Many quantitative trait loci (QTL) for PHS resistance have been reported in different sources. To determine the genetic architecture of PHS resistance and its relationship with grain color (GC) in US hard winter wheat, a genome-wide association study (GWAS) on both PHS resistance and GC was conducted using in a panel of 185 U.S. elite breeding lines and cultivars and 90K wheat SNP arrrays. PHS resistance was assessed by evaluating sprouting rates in wheat spikes harvested from both greenhouse and field experiments. Thirteen QTLs for PHS resistance were identified on 11 chromosomes in at least two experiments, and the effects of these QTLs varied among different environments. The common QTLs for PHS resistance and GC were identified on the long arms of the chromosome 3A and 3D, indicating pleiotropic effect of the two QTLs. Significant QTLs were also detected on chromosome arms 3AS and 4AL, which were not related to GC, suggesting that it is possible to improve PHS resistance in white wheat. To identify markers closely linked to the 4AL QTL, genotyping-by-sequencing (GBS) technology was used to analyze a population of recombinant inbred lines (RILs) developed from a cross between two parents, “Tutoumai A” and “Siyang 936”, contrasting in 4AL QTL. Several closely linked GBS SNP markers to the 4AL QTL were identified and some of them were coverted to KASP for marker-assisted breeding. To investigate effects of the two non-GC related QTLs on 3AS and 4AL, both QTLs were transferered from “Tutoumai A” and “AUS1408” into a susceptible US hard winter wheat breeding line, NW97S186, through marker-assisted backcrossing using the gene marker TaPHS1 for 3AS QTL and a tightly linked KASP marker we developed for 4AL QTL. The 3AS QTL (TaPHS1) significantly interacted with environments and genetic backgrounds, whereas 4AL QTL (TaMKK3-A) interacted with environments only. The two QTLs showed additive effects on PHS resistance, indicating pyramiding these two QTLs can increase PHS resistance. To improve breeding selection efficiency, genomic prediction using genome-wide markers and marker-based prediction (MBP) using selected trait-linked markers were conducted in the association panel. Among the four genomic prediction methods evaluated, the ridge regression best linear unbiased prediction (rrBLUP) provides the best prediction among the tested methods (rrBLUP, BayesB, BayesC and BayesC0). However, MBP using 11 significant SNPs identified in the association study provides a better prediction than genomic prediction. Therefore, for traits that are controlled by a few major QTLs, MBP may be more effective than genomic selection.

Master of Science / Department of Statistics / Haiyan Wang / Motivation: Genome-wide association studies (GWAS) provide an important avenue for investigating many common genetic variants in different individuals to see if any variant is associated with a trait. GWAS is a great tool to identify genetic factors that influence health and disease. However, the high dimensionality of the gene expression dataset makes GWAS challenging. Although a lot of promising machine learning methods, such as Support Vector Machine (SVM), have been investigated in GWAS, the question of how to improve the accuracy of the result has drawn increased attention of many researchers A lot of the studies did not apply feature selection to select a parsimonious set of relevant genes. For those that performed gene selections, they often failed to consider the possible interactions among genes. Here we modify a gene selection algorithm BMSF originally developed by Zhang et al. (2012) for improving the accuracy of cancer classification with binary responses. A continuous response version of BMSF algorithm is provided in this report so that it can be applied to perform gene selection for continuous gene expression dataset. The algorithm dramatically reduces the dimension of the gene markers under concern, thus increases the efficiency and accuracy of GWAS. Results: We applied the continuous response version of BMSF on the wheat phenotypes dataset to predict two quantitative traits based on the genotype marker data. This wheat dataset was previously studied in Long et al. (2009) for the same purpose but used only direct application of SVM regression methods. By applying our gene selection method, we filtered out a large portion of genes which are less relevant and achieved a better prediction result for the test data by building SVM regression model using only selected genes on the training data. We also applied our algorithm on simulated datasets which was generated following the setting of an example in Fan et al. (2011). The continuous response version of BMSF showed good ability to identify active variables hidden among high dimensional irrelevant variables. In comparison to the smoothing based methods in Fan et al. (2011), our method has the advantage of no ambiguity due to difference choices of the smoothing parameter.

The main objective of the present thesis was to propose a procedure to optimize genotypic information value in small dairy cattle populations and investigate the impacts of including genotypes and phenotypes of cows chosen by different strategies over the performance of genome-wide association studies and genomic selection. The first study was designed to propose innovative methods that could support alternative inference over population structure in livestock populations using graph theory. It reviews general aspects of graphs and how each element relates to theoretical and practical concepts of traditional pedigree structure studies. This chapter also presents a computational application (PedWorks) built in Python 2.7 programming language. It demonstrates that graph theory is a suitable framework for modeling pedigree data. The second study was aimed asses how graph community detection algorithms could help unraveling population partition. This new concept was considered to develop a method for stablishing new cow genotyping strategies (community-based). Results obtained showed that accounting for population structure using community detection for choosing cows to get included in the reference population may improve results from genomic selection. Methods presented are easily applied to animal breeding programs. The third study aimed to observe the impacts of different genotyping strategies (including the proposed community-based) over the ability to detect quantitative trait loci in genome-wide association studies. Distinct models for genomic analysis were also tested. Results obtained showed that including cows with extreme phenotypic observations proportionally sampled from communities can improve the ability to detect quantitative trait loci in genomic evaluations. The last chapter was designed study possible deleterious impacts of the presence of preferential treatment (in different levels) in a small dairy cattle population environment over accuracy and bias of genomic selection. Different proportions of cows with artificially increased phenotypic observations were included in the reference population. Observed results suggest that both accuracy and bias are affected by the presence of preferential treatment of cows in the evaluated population. Preferential treatment is expected to have much more effect on the performance of genomic selection in small than in large dairy cattle populations for the higher (proportional) value of the information from cows in such reduced-size breeds. / O principal objetivo da presente tese foi propor um procedimento capaz de otimizar o valor da informação genotípica em pequenas populações de gado de leite e investigar os impactos da inclusão de genótipos e fenótipos de vacas escolhidas por diferentes estratégias sobre o desempenho de estudos de associação genômica ampla e seleção genômica. O primeiro estudo foi delineado para elaborar um método que permita uma inferência alternativa sobre a estrutura populacional de populações de animais de produção usando como base a teoria de grafos. Este revê os aspectos gerais de grafos e como cada elemento se relaciona com conceitos teóricos e práticos de estudos de estrutura de pedigree tradicionais. Este capítulo também apresenta um aplicativo computacional (PedWorks) construído em linguagem de programação Python 2.7. Resultados observados demonstraram que a teoria de grafos é uma estrutura adequada para modelar dados de pedigree. O segundo estudo teve como objetivo avaliar como os algoritmos de detecção de comunidades de grafos poderiam ajudar revelar o particionamento de uma população. Este novo conceito foi considerado para desenvolver um método para o estabelecimento de novas estratégias de genotipagem de vacas (baseadas em comunidades). Os resultados obtidos mostraram que a contabilização da estrutura populacional usando a detecção de comunidades para a escolha de vacas a serem incluídas na população de referência pode melhorar os resultados da seleção genômica. Os métodos apresentados sugerem ser facilmente introduzidos em programas de melhoramento animal. O terceiro estudo teve como objetivo observar os impactos de diferentes estratégias de genotipagem (incluindo a anteriormente proposta baseada em comunidades) sobre a capacidade de detectar locos relacionados características quantitativas por meio de estudos de associação genômica ampla. Modelos distintos para análise genômica também foram testados. Os resultados obtidos mostraram que incluir vacas com observações fenotípicas extremas amostradas proporcionalmente das comunidades pode melhorar a capacidade de detectar locos de características quantitativas em avaliações genômicas. O último capítulo foi desenhado para estudar possíveis impactos deletérios da presença de tratamento preferencial no ambiente de pequenas populações de gado leiteiro sobre resultados da seleção genômica. Diferentes proporções de vacas com observações fenotípicas aumentadas artificialmente foram incluídas na população de referência. Os resultados observados sugerem que tanto a acurácia quanto o viés são afetados pela presença de tratamento preferencial de vacas na população avaliada. Espera-se que o tratamento preferencial tenha muito mais efeito sobre o desempenho da seleção genômica em populações pequenas de gado de leite que em grandes populações devido a maior relevância das informações de vacas em raças de tamanho reduzido.

The Nelore breed represents the vast majority of Brazilian Zebuine cattle (Bos taurus indicus). The great adaptability of the Nelore breed to Brazilian tropical climate, however, is not associated with meat tenderness (MT). It is known that MT is influenced by several environmental factors, but also genetic composition. In the first chapter, we report a genome-wide analysis of copy number variation (CNV) inferred from Illumina® Bovine High Density SNP-chip data for a Nelore population of 723 males including 30 sires. We detected >2600 CNV regions (CNVRs) representing &asymp;6.5% of the Bos taurus genome. The CNVR size was 65 kb on average, ranging from 5 kb to 4.3 Mb. A total of 1155 CNVRs (43.6%) overlapped 2750 genes. They are enriched for important functions such as immune response, olfactory reception and processes involving guanosine triphosphate (GTP). The GTP processes have known influence in skeletal muscle physiology and morphology. Quantitative trait loci for MT, partly specific for Nelore, overlapped a substantial fraction of CNVRs and two CNVRs were found proximal to glutathione metabolism genes that are associated with MT as well. Comparing our results with previous studies revealed an overlap in &asymp;1400 CNVRs (>50%). We selected 9 CNVRs that overlapped regions associated with MT and we validated them in all 30 sires by qPCR. There was identified many genomic regions of structural variation in Nelore with important implications on the MT phenotype. In the second chapter, a total of 34 animals of the population were subjected to transcriptome analysis and meat tenderness (MT) phenotyping. We identified 170 CNV fragments (CNVFs) residing in 20 CNVRs, which occurred in different frequencies between animals with tougher and softer meat genetic potential. A considerable fraction of the identified CNVFs affected gene expression of the MT genes, which play important roles in glycogen metabolism, connective tissue turnover, membrane transporters and glutathione pathways. We also detected that several CNVRs substantially influenced the expression of overlapped and nearby genes, where the increase or decrease of copy number correlated well with the change in gene expression. Among them are two CNVRs at chromosomes 12 and 23, which are in the vicinity of previously described QTLs for MT in Nelore breed. Several CNVFs, which are more frequent in animals with genetic potential for softer or tougher MT, showed significant differences in gene expression. Those regions are linked to important biological functions with highly relevant influences on MT and skeletal muscle physiology. / A raça Nelore é predominante no rebanho zebuíno brasileiro (Bos taurus indicus). A grande adaptabilidade da raça Nelore ao clima tropical brasileiro, no entanto, não está associada à maciez de carne (MT). Sabe-se que MT é influenciada por vários fatores ambientais e pela composição genética. Foi realizada uma análise de todo o genoma para inferir Variação no Número de Cópias de Segmentos Genômicos (Copy Number Variation - CNV) a partir de dados oriundos de chip de SNP (Illumina® Bovine High Density), para uma população de 723 machos Nelore, incluindo 30 ancentrais da população. Foram detectadas >2600 regiões de CNV (CNVRs) representando &asymp;6.5% do genoma bovino. O tamanho médio do CNVR foi de 65 kb, variando de 5 kb até 43 Mb. Um total de 1155 CNVRs (43.6%) obtiveram sobreposição com 2750 genes. Estes genes foram enriquecidos para as funções importantes, tais como resposta imunológica, recepção olfativa e processos que envolvem o trifosfato de guanosina (GTP). As vias metabólicas do GTP conhecidamente influenciam a fisiologia e a morfologia do músculo esquelético. Loci de características quantitativas (QTLs) para MT, alguns específicos para Nelore, sobrepuseram uma fração substancial das CNVRs encontradas. Dois CNVRs foram encontrados em região proximal à genes do metabolismo da glutationa os quais também são associados com MT. Comparando os resultados com estudos anteriores &asymp;1400 CNVRs (>50%) foram sobrepostos. Nove CNVRs em regiões associadas com MT foram validados nos 30 ancentrais por qPCR. Em conclusão, foram identificadas regiões genômicas de variação estrutural no Nelore, com potenciais implicações sobre o fenótipo MT. No segundo capítulo, um total de 34 animais da população foi submetido à análise do transcriptoma e análise de potencial genético para MT. Foram identificados 170 fragmentos de CNV (CNVFs) mapeados em 20 CNVRs, os quais mostraram frequências significativamente diferentes entre animais com potencial genético para carne mais dura ou mais macia. Uma fração considerável dos CNVFs identificados afetaram a expressão gênica de genes MT (anteriormente descritos como associados à MT ou fisiologia do músculo esquelético), os quais desempenham um papel importante no metabolismo de glicogênio, volume do tecido conjuntivo, transportadores de membrana e vias metabólicas da glutationa. Um número considerável de CNVRs foram associados à expressão de genes sobrepostos e nas proximidades, onde o aumento ou diminuição do número de cópias foi associado com a mudança na expressão gênica. Dois CNVRs associados foram mapeados para os cromossomo 12 e 23, estando próximos a QTLs anteriormente descritos para MT na raça Nelore. Vários CNVFs, entre animais com potencial genético para carne mais macia ou dura, mostraram diferenças significativas na expressão gênica. Essas regiões estão ligadas a importantes funções biológicas com influências altamente relevantes para MT e para a fisiologia do músculo esquelético.

Parkinson's Disease (PD) is the second most common neurodegenerative disorder, clinically characterized by severe motor impairment and pathologically characterized by progressive loss of the dopaminergic neurons of the substantia nigra pars compacta (SNpc) as well as the formation of cellular aggregate deposits called Lewy Bodies. While some advances have been made in understanding the molecular underpinnings of the disorder, the molecular implications of common genetic factors increasing risk for PD have not been adequately studied. First identified by GWA studies in 2009, the GAK/DGKQ/IDUA region on chromosome 4p16.3 shows significant genetic association to risk for PD, and the GAK protein has been shown to be associated with the primary component of Lewy Bodies, a-synuclein. In order to determine which gene in the 4p16.3 region may account for the genetic association to PD and to understand the molecular consequences of that association, post-mortem cortical brain tissue from 29 PD and 49 control patients was RNA-sequenced and differential exon usage in the context of disease and risk variant carrier status was analyzed. Exons in the 3' region of GAK were found to be associated to case status, and notably exon 25 expression in GAK was associated with both case status and the risk variant. This exon was further observed to be associated to several genes previously shown to interact with GAK, including SNCA, which codes for a-synuclein. As a proxy for expression of the 3' region of GAK, exon 25 was assessed for genome-wide association, and genes showing association to the exon were involved in pathways related to synaptic transmission and neuronal function. In order to validate these findings, microarray analysis of primary rat cortical neurons in which GAK expression was reduced by shRNA transduction was performed. GAK expression in rat neurons was significantly inversely correlated to endogenous SNCA expression, and also exhibited association to pathways involved in synaptic transmission and mitochondrial function. Together, these findings suggest aberrant GAK expression related to genetic risk to be an important factor in the pathogenesis of PD through GAK's influence on SNCA expression and through dysregulation of important neuronal pathways.

The publication of a draft equine genome sequence and the release by Illumina of a 50,000 marker single-nucleotide polymorphism (SNP) genotyping chip has provided equine researchers with the opportunity to use new approaches to study the relationships between genotype and phenotype. In particular, it is hoped that the use of high-density markers applied to population samples will enable progress to be made with regard to more complex diseases. The first objective of this thesis is to explore the potential for the equine SNP chip to enable such studies to be performed in the horse. The second objective is to investigate the genetic background of osteochondrosis (OC) in the horse. These objectives have been tackled using 348 Thoroughbreds from the US, divided into cases and controls, and a further 836 UK Thoroughbreds, the majority with no phenotype data. All horses had been genotyped with the Illumina Equine SNP50 BeadChip. Linkage disequilibrium (LD) is the non-random association of alleles at neighbouring loci. The reliance of many genomic methodologies on LD between neutral markers and causal variants makes it an important characteristic of genome structure. In this thesis, the genomic data has been used to study the extent of LD in the Thoroughbred and the results considered in terms of genome coverage. Results suggest that the SNP chip offers good coverage of the genome. Published theoretical relationships between LD and historical effective population size (Ne) were exploited to enable accuracy predictions for genome-wide evaluation (GWE) to be made. A subsequent in-depth exploration of this theory cast some doubt on the reliability of this approach in the estimation of Ne, but the general conclusion that the Thoroughbred population has a small Ne which should enable GWE to be carried out efficiently in this population, remains valid. In the course of these studies, possible errors embedded within the current sequence assembly were identified using empirical approaches. Osteochondrosis is a developmental orthopaedic disease which affects the joints of young horses. Osteochondrosis is considered multifactorial in origin with a variety of environmental factors and heredity having been implicated. In this thesis, a genome-wide association study was carried out to identify quantitative trait loci (QTL) associated with OC. A single SNP was found to be significantly associated with OC. The low heritability of OC combined with the apparent lack of major QTL suggests GWE as an alternative approach to tackle this disease. A GWE analysis was carried out on the same dataset but the resulting genomic breeding values had no predictive ability for OC status. This, combined with the small number of significant QTL, indicates a lack of power which could be addressed in the future by increasing sample size. An alternative to genotyping more horses for the 50K SNP chip would be to use a low-density SNP panel and impute remaining genotypes. The final chapter of this thesis examines the feasibility of this approach in the Thoroughbred. Results suggest that genotyping only a subset of samples at high density and the remainder at lower density could be an effective strategy to enable greater progress to be made in the arena of equine genomics. Finally, this thesis provides an outlook on the future for genomics in the horse.

Objective Genetic factors play an important role in modulating the vulnerability to body mass index (BMI). The purpose of this study is to identify novel genetic variants for BMI using genome-wide association (GWA) meta-analysis. Methods PLINK software was used to perform meta-analysis of two GWA studies (the FUSION and Marshfield samples) of 5218 Caucasian individuals with BMI. A replication study was conducted using the SAGE sample with 762 individuals. Results Through meta-analysis we identified 33 SNPs associated with BMI with p < 10− 4. The most significant association was observed with rs2967951 (p = 1.19 × 10− 6) at 5p15.2 within ROPN1L gene. Two additional SNPs within ROPN1L and 5 SNPs within MARCH6 (the top SNP was rs2607292 with 4.27 × 10− 6) further supported the association with BMI on 5p15.2 (p < 1.8 × 10− 5). Conditional analysis on 5p15.2 could not distinguish the effects of ROPN1L and MARCH6. Several SNPs within MARCH6 and ROPN1L were replicated in the SAGE sample (p < 0.05). Conclusion We identified a novel locus for BMI. These findings offer the potential for new insights into the pathogenesis of BMI and obesity and will serve as a resource for replication in other populations to elucidate the potential role of these genetic variants in BMI and obesity.

M.S. / Computer Science and Engineering / Genome-wide studies are sensitive to the quality of annotation data included for analyses and they often involve overlaying both computationally derived and experimentally generated data onto a genomic scaffold. A framework for successful integration of data from diverse sources needs to address, at a minimum, the conceptualization of the biological identity in the data sources, the relationship between the sources in terms of the data present, the independence of the sources and, any discrepancies in the data. The outcome of the process should either resolve or incorporate these discrepancies into downstream analyses. In this thesis we identify factors that are important in detecting errors within and between sources and present a generalized framework to detect discrepancies. An implementation of our workflow is used to demonstrate the utility of the approach in the construction of a genome-wide mouse transcription factor binding map and in the classification of Single nucleotide polymorphisms. We also present the impact of these discrepancies on downstream analyses. The framework is extensible and we discuss future directions including summarization of the discrepancies in a biological relevant manner.

Systemic Lupus Erythematosus (SLE) is a chronic multi-organ autoimmune disease considered a prototype for autoantibody and immune complex-mediated tissue injury. Although autoantibodies against a wide diversity of self-antigens are characteristically found in this disease, an important hallmark is the presence of autoantibodies to nuclear antigens. Despite this common clinical feature, individual patients vary widely in the organ systems afflicted, disease severity, disease course, and response to treatment. These characteristics make clinical management of SLE challenging and highlight the need for effective and less toxic therapeutic interventions. Susceptibility to lupus has been shown in both human studies and mouse models to be dependent on genetic predisposition. Therefore, it is likely that knowledge of the genetic basis of SLE will be required before full understanding of SLE pathogenesis can be achieved. In this thesis, studies to define the genetic basis of lupus in an induced and two spontaneous models of the disease are presented. These studies encompass mapping, characterization of interval congenic mice, and cloning of the Lmb3 locus gene. In the first study, a genomewide mapping study was performed to define the genetic basis for resistance of the DBA/2 mice to mercury-induced autoimmunity. On chromosome 1, a single quantitative trait was linked with resistance to HgIA. These results linked the locus Hmr1 to a late stage of lupus with GN. Interval congenic mice are important tools to define and characterize the roles of different loci in lupus-like diseases. The second paper identifies the effect of NZB and NZW Lbw2 alleles on lupus susceptibility by using BWF1 mice with none, one or two copies of the lupus-predisposing NZB.Lbw2 locus. The lack of the NZB locus significantly reduced mortality, GN and B cell activation. IgM anti-chromatin levels in genome-wide mapping was linked only to Lmb2 and none of the known B cell hyperactivity-promoting genes were present in this location, which might indicate a novel B cell activation gene. The third study used reciprocal single locus interval-specific congenic mice to characterize the contribution of Lmb1-4 on the MRL-Faslpr and B6-Faslpr backgrounds. The Lmb3 locus on chromosome 7 was found to have the most prominent phenotype with clear effects on lymphoproliferation, GN and mortality. In the fourth paper the Lmb3 was cloned and shown to be a spontaneous nonsense mutation in the Coro1a gene that encodes an actin-binding and -regulatory protein. Upon further characterization, this genetic alteration was discovered to be a new lupus suppressing mutation that reduced T cell migration, activation, and survival. Our findings highlight the complexity of the genetics of lupus, and further suggest that genes involved in controlling the actin cytoskeleton might be potential targets for autoimmune therapeutics.