Genome-wide association studies (GWASs) have reported numerous associations between risk variants and Alzheimer’s disease (AD). However, these associations do not necessarily indicate a causal relationship. If the risk variants can be demonstrated to be biologically functional, the possibility of a causal relationship would be increased. In this article, we reviewed all of the published GWASs to extract the genome-wide significant (p < 5×10−8) and replicated associations between risk variants and AD or AD-biomarkers. The regulatory effects of these risk variants on the expression of a novel class of non-coding RNAs (piRNAs) and protein-coding RNAs (mRNAs), the alteration of proteins caused by these variants, the associations between AD and these variants in our own sample, the expression of piRNAs, mRNAs and proteins in human brains targeted by these variants, the expression correlations between the risk genes and APOE, the pathways and networks that the risk genes belonged to, and the possible long non-coding RNAs (LncRNAs) that might regulate the risk genes were analyzed, to investigate the potential biological functions of the risk variants and explore the potential mechanisms underlying the SNP-AD associations. We found replicated and significant associations for AD or AD-biomarkers, surprisingly, only at 17 SNPs located in 11 genes/snRNAs/LncRNAs in eight genomic regions. Most of these 17 SNPs enriched some AD-related pathways or networks, and were potentially functional in regulating piRNAs and mRNAs; some SNPs were associated with AD in our sample, and some SNPs altered protein structures. Most of the protein-coding genes regulated by the risk SNPs were expressed in human brain and correlated with APOE expression. We conclude that these variants were most robust risk markers for AD, and their contributions to AD risk was likely to be causal. As expected, APOE and the lipoprotein metabolism pathway possess the highest weight among these contributions.

A comprehensive understanding about how genetic variants and mutations contribute to phenotypic variations and alterations entails experimental technologies and analytical methodologies that are able to detect genetic variants/mutations from various biological samples in a timely and accurate manner. High-throughput sequencing technology represents the latest achievement in a series of efforts to facilitate genetic variants discovery and genotyping and promises to transform the way we tackle healthcare and biomedical problems. The tremendous amount of data generated by this new technology, however, needs to be processed and analyzed in an accurate and efficient way in order to fully harness its potential. Structural variation (SV) encompasses a wide range of genetic variations with different sizes and generated by diverse mechanisms. Due to the technical difficulties of reliably detecting SVs, their characterization lags behind that of SNPs and indels. In this dissertation I presented two novel computational methods: one for detecting transposable element (TE) transpositions and the other for detecting SVs in general using a local assembly approach. Both methods are able to pinpoint breakpoint junctions at single-nucleotide resolution and estimate variant allele frequencies in the sample. I also applied those methods to study the impact of TE transpositions on the genomic stability, the inheritance patterns of TE insertions in the population and the molecular mechanisms and potential functional consequences of somatic SVs in cancer genomes.

京都大学 / 0048 / 新制・課程博士 / 博士(医学) / 甲第19402号 / 医博第4053号 / 新制||医||1012(附属図書館) / 32427 / 京都大学大学院医学研究科医学専攻 / (主査)教授 佐藤 俊哉, 教授 小川 誠司, 教授 横出 正之 / 学位規則第4条第1項該当 / Doctor of Medical Science / Kyoto University / DFAM

京都大学 / 0048 / 新制・課程博士 / 博士(医学) / 甲第20278号 / 医博第4237号 / 新制||医||1021(附属図書館) / 京都大学大学院医学研究科医学専攻 / (主査)教授 森田 智視, 教授 佐藤 俊哉, 教授 中山 健夫 / 学位規則第4条第1項該当 / Doctor of Medical Science / Kyoto University / DFAM

No description available.

Here we present the genome-wide association study of body weight at 8 weeks of age based onthe advanced intercross pedigree of two chicken lines gone through bi-directional selection.With improved marker density (~3M SNPs) and larger sample size (2667 individuals from F2-F15), 34 loci with suggestive significance are detected, of which 18 loci are novel, and the rest17 loci are consistent with the results of previous quantitative trait locus mapping studies onthis trait with smaller number of genetic markers and fewer individuals. The component traits,referring to traits related to body weight and possibly contributing to the body weight as well,are also measured and analysed. The combined result showed that one locus with significantmarginal effect on BW8 is associated with early growth, breast muscle development and shankdevelopment, while another locus with late development and bursa development.

Brain metastasis (BM), the most common tumor of the central nervous system, occurs in 20-36% of primary cancers. In particular, 20-40% of patients with non-small cell lung cancer (NSCLC) develop brain metastases, with a dismal survival of approximately 4-11 weeks without treatment, and 16 months with treatment. This highlights a large unmet need to develop novel targeted therapies for the treatment of lung-to-brain metastases (LBM). Genomic interrogation of LBM using CRISPR technology can inform preventative therapies targeting genetic vulnerabilities in both primary and metastatic tumors. Loss-of-function studies present limitations in metastasis research, as knocking out genes essential for survival in the primary tumor cells can thwart the metastatic cascade prematurely. However, transcriptional overexpression of genes using CRISPR activation (CRISPRa) has the potential for overcoming dependencies of gene essentiality. In this thesis, we created and utilized an in vivo genome-wide CRISPRa screening platform to identify novel genes, that when overexpressed, drive LBM. We have developed a patient-derived orthotopic murine xenograft model of LBM using a patient-derived NSCLC cell line (termed CRUK cells) from the Swanton Lab TRACERx study. We introduced a human genome-wide CRISPRa single guide RNA (sgRNA) library into non-metastatic and pro-metastatic lung cancer CRUK cells to achieve 500X representation of each sgRNA in the activation library. We then injected the cells into the lungs of immunocompromised mice and tracked lung tumor development and BM formation. Upon sequencing primary lung tumors and subsequent BM, we will identify enriched sgRNAs which may represent novel drivers of primary lung tumor formation and LBM. To the best of our knowledge, this study is the first in vivo genome-wide CRISPR activation screen using patient-derived NSCLC cells to help elucidate drivers of LBM. This work serves to provide a framework to gain a deeper understanding of the regulators of BM formation which will hopefully lead to targeted drug discovery that will ultimately be used in clinical trials to help eradicate brain metastasis in NSCLC patients. / Thesis / Master of Science (MSc) / Brain metastasis, or the spread of a primary cancer from another organ to the brain, is the most common adult brain tumor. Brain metastases can arise after the treatment of primary tumors and are only detected in the clinic at a highly malignant stage. Current treatments for brain metastasis consist of surgical removal and palliative chemoradiotherapy, which fail to fully eliminate the brain tumor. Over 20% of cancer patients develop brain metastases, with lung, breast, and skin cancers leading as the top three sources of metastasis. In particular, 40% of patients with non-small cell lung cancer develop brain metastasis, with survival of only 4-11 weeks once diagnosed without treatment, and 16 months with treatment. As systemic therapies for the treatment of non-small cell lung cancer are becoming increasingly effective at controlling primary disease, patients are ironically succumbing to their brain tumors. This highlights a large unmet need to develop novel targeted therapies for the treatment of lung-to-brain metastases (LBM). Functional genomic tools provide the opportunity to investigate the genetic underpinnings of LBM. With the advent of gene editing technologies, we are able to overexpress various genes and observe the impact genetic perturbations have on tumor initiation, growth, and metastasis. In this thesis, we devised a pre-clinical animal model of LBM that could be used to study genetic drivers of LBM using a gene overexpression tool such that one gene per tumor cell gets activated. We are then able to model the disease trajectory from a lung tumor to brain metastasis development using patient samples in our animal model and identify genes that, upon overexpression, drive LBM. This platform will lead to potential therapeutic targets to prevent the formation of LBM and prolong the survival of patients with non-small cell lung cancer.

The Norway spruce is of great importance from both an ecological- and economic standpoint. Information about which genes that causes certain phenotypic traits in the species is therefore highly valuable. The purpose of this project was to apply machine learning to find such genotype-phenotype correlations. The purpose was also to compare the results from different machine learning algorithms to a more traditional linear mixed model GWAS (where correlation to the phenotype is estimated for each SNP one by one) to find which is the better method for GWAS. The machine learning algorithms tested were decision tree, support vector machine and support vector regression. The phenotypes analyzed were wood density and initiation frequency of zygotic embryogenesis (ZE). The latter is related to a new method for cloning. The genetic data consisted of single-nucleotide polymorphisms (SNPs). Due to the large genome size of Norway spruce and due to limitations in the packages used in R two different approaches were taken to reduce the sample size. The first approach used Kendall’s rank correlation coefficient to remove redundant SNPs and the second used an iterative approach to the machine learning model. The iterative approach was proven to be the best and support vector machine/regression was found to be better than decision tree for both phenotypes. Support vector regression from the iterative approach resulted in a squared correlation coefficient of 0.83 for density and 0.94 for ZE initiation frequency. Note that these very high values should be interpreted with caution, as it is possible that some of the significant correlations are only due to random chance. Even a small chance for random correlations will result in findings when the number of SNPs are this large (1908552 SNPs). The significant SNPs identified by the machine learning models were compared to SNPs identified by the linear mixed model GWAS. This indeed showed some overlaps of significant SNPs, which increases the credibility of my results. However, further investigation of the identified significant SNPs is needed to determine their functional mode of action. My conclusion is that using machine learning to predict phenotypic traits from SNP data can be a good choice. However, the model might not use all correlated SNPs, just enough to get a good prediction. Therefore, for the purpose of finding significant SNPs, the linear mixed model approach might be better. In other words, the method used should be determined by the purpose of the study.

京都大学 / 新制・課程博士 / 博士(医学) / 甲第24494号 / 医博第4936号 / 新制||医||1063(附属図書館) / 京都大学大学院医学研究科医学専攻 / (主査)教授 村川 泰裕, 教授 小杉 眞司, 教授 松田 文彦 / 学位規則第4条第1項該当 / Doctor of Medical Science / Kyoto University / DFAM

Bipolar disorder is a relatively common form of mental illness that depends strongly on genetic inheritance for expression. The author of this study has sought to evaluate whether the gender of subjects influences which genetic variants are associated with the disease. A portion of the cases from a previously published study were analyzed using PLINK software and the association of single-nucleotide polymorphisms was evaluated separately for all cases, for female subjects alone, and for male subjects alone. The results obtained for male subjects alone reached higher levels of statistical significance than when both genders were evaluated together or when female subjects were evaluated alone. The most significantly scoring polymorphisms were distinctly different for the 2 genders. In particular, a site downstream of the ion exchanger SLC24A3 and upstream of the Rab5-interacting protein RIN2 gene on chromosome 20 (rs6046396) yielded very high significance in men (p=3.91 X 10-9).