Single-cell RNA sequencing and in situ sequencing were combined in a recent study of the developing human heart to explore the transcriptional landscape at three developmental stages. However, the method used in the study to create the spatial cellular maps has some limitations. It relies on image segmentation of the nuclei and cell types defined in advance by single-cell sequencing. In this study, we applied a new unsupervised approach based on graph neural networks on the in situ sequencing data of the human heart to find spatial gene expression patterns and detect novel cell and sub-cell types. In this thesis, we first introduce some relevant background knowledge about the sequencing techniques that generate our data, machine learning in single-cell analysis, and deep learning on graphs. We have explored several graph neural network models and algorithms to learn embeddings for spatial gene expression. Dimensionality reduction and cluster analysis were performed on the embeddings for visualization and identification of biologically functional domains. Based on the cluster gene expression profiles, locations of the clusters in the heart sections, and comparison with cell types defined in the previous study, the results of our experiments demonstrate that graph neural networks can learn meaningful representations of spatial gene expression in the human heart. We hope further validations of our clustering results could give new insights into cell development and differentiation processes of the human heart.

京都大学 / 新制・課程博士 / 博士(医学) / 甲第23082号 / 医博第4709号 / 新制||医||1049(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 長船 健二, 教授 妹尾 浩, 教授 村川 泰裕 / 学位規則第4条第1項該当 / Doctor of Medical Science / Kyoto University / DFAM

京都大学 / 新制・課程博士 / 博士(医学) / 甲第23382号 / 医博第4751号 / 新制||医||1052(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 伊藤 貴浩, 教授 滝田 順子, 教授 江藤 浩之 / 学位規則第4条第1項該当 / Doctor of Medical Science / Kyoto University / DFAM

The definitive endoderm gives rise to several specialized organs, including the thymus. Improper development of the definite endoderm or its derivatives can lead to human disease; in the case of the thymus, immunodeficiency or autoimmune disorders. Human pluripotent stem cells (hPSCs) have emerged as a system to model human development, as study of their differentiation allows for elucidation of the molecular basis of cell fate decisions, under both healthy and impaired conditions. Here, we first developed a CRISPR-effector system to control endogenous gene expression in hPSCs, a novel approach to manipulating hPSC state. Next, the human-specific, loss-of-function phenotypes of candidate transcription factors driving hPSC-to-definitive endoderm differentiation were analyzed through combined CRISPR-perturbation and single-cell RNA-sequencing. This analysis revealed the importance of TGFβ mediators in human definitive endoderm differentiation as well as identified an unappreciated role for FOXA2 in human foregut development. Finally, as the differentiation of definitive endoderm to thymic epithelial progenitors (TEPs) is of particular interest, a single-cell transcriptomic atlas of murine thymus development was generated in anticipation of identifying factors driving later stages of TEP differentiation. Taken together, this dissertation establishes a CRISPR-effector system to interrogate gene and regulatory element function in hPSC differentiation strategies, details the role of specific transcription factors in human endoderm differentiation, and sets the groundwork for future investigations to characterize hPSC-derived TEPs and the factors driving their differentiation.

No description available.

Die Alzheimer-Krankheit ist eine gewaltige Bedrohung für eine alternde Gesellschaft. Millionen von Menschen leben weltweit mit der Alzheimer-Krankheit, für die es keine aktuelle Behandlung gibt. Die Amyloidkaskaden-Hypothese (AKH) ist die aktuell am meisten akzeptierte Hypothese zur Ursache der Alzheimer-Krankheit. Die AKH bietet eine mechanistische Sicht auf die pathologische Kaskade, ausgehend von der Amyloid-Aggregation über die chronische Entzündung bis hin zur TAU-Pathologie. Die Medikamente, die auf der Grundlage der AKH entwickelt wurden, konnten Amyloid-Plaques bei Alzheimer-Patienten entfernen, brachten aber keine Verbesserung der kognitiven Fähigkeiten. Diese Misserfolge legen nahe, dass die Alzheimer-Krankheit nicht nur theoretisch im Rahmen der AKH betrachtet werden kann. Neuere Hypothesen kulminieren die Auswirkungen verschiedener Zelltypen (z.B. neurale Stammzellen, Astrozyten, Oligodendrozyten) auf den Ausbruch der Alzheimer-Erkrankung. Komplexe Rückkopplungs- und Feed-Forward-Mechanismen sind in der Pathophysiologie der Alzheimer-Demenz wahrscheinlich. Das Zusammenspiel zwischen der Pathologie und der Beteiligung anderer Zelltypen macht diese Krankheit multifaktoriell und komplex. Kürzlich zeigten zwei Studien (Moreno-Jimenez et al., 2019; Tobin et al., 2019), dass die Produktion neuer Neuronen im menschlichen Gehirn bei der Alzheimer-Erkrankung dramatisch abnimmt. Eine interessante Hypothese wurde durch diese Studien gestützt: Die pathologisch induzierte Erzeugung neuer Neuronen (regenerative Neurogenese) bei Alzheimer-Patienten könnte helfen, die Pathologie der Alzheimer-Erkrankung rückgängig zu machen. Da die Regenerationsfähigkeit bei Säugetieren entwicklungsmäßig wenig ausgeprägt ist (Tanaka und Ferretti, 2009), kann uns die Untersuchung der Neurodegeneration in einem Modellorganismus mit Regenerationsfähigkeit daher lehren, wie man die Proliferation und Neurogenese neuraler Stammzellen unter pathologischen Bedingungen induzieren kann. Für diese spezielle Frage können uns Modellorganismen mit natürlicher Regenerationsfähigkeit zeigen, wie man Proliferation und Neurogenese unter den pathologischen Bedingungen der Alzheimer-Erkrankung induzieren kann. Der Zebrafisch bietet eine beispiellose Möglichkeit, die Neurodegeneration und Regeneration zu modellieren, um die molekularen Mechanismen zu untersuchen, wie anhand der Neurogenese in Wirbeltiergehirnen die Alzheimer-Krankheit verbessert werden kann. Dies wurde in unserem Labor bereits in mehreren Publikationen gezeigt. Aus diesem Grund habe ich in meiner Doktorarbeit Zebrafische verwendet, um die Plastizität neuraler Stammzellen (NSZ) zu untersuchen. Besonders interessierte mich die Heterogenität von NSZ-Populationen in Bezug auf ihre molekularen Programme und die molekulare Grundlage der regenerativen Neurogenese von NSZ auf das Amyloid-β-42 (Aβ42) und TAU-Pathologien.

京都大学 / 0048 / 新制・課程博士 / 博士(医学) / 甲第19551号 / 医博第4058号 / 新制||医||1012(附属図書館) / 32587 / 京都大学大学院医学研究科医学専攻 / (主査)教授 川口 義弥, 教授 坂井 義治, 教授 羽賀 博典 / 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(農学) / 甲第19759号 / 農博第2155号 / 新制||農||1039(附属図書館) / 学位論文||H28||N4975(農学部図書室) / 32795 / 京都大学大学院農学研究科応用生命科学専攻 / (主査)教授 植田 充美, 教授 宮川 恒, 教授 栗原 達夫 / 学位規則第4条第1項該当 / Doctor of Agricultural Science / Kyoto University / DFAM

Technological advancement allowed for the advent of single-cell technologies capable of measuring a large number of cellular features simultaneously. These technologies have been subsequently used to shed light on the heterogeneity of cellular systems previously considered homogeneous, identifying the exclusive features of individual cells within cellular niches. Today, single-cell technologies represent an essential tool for studying the underlying immunological mechanisms correlating with disease. In this context, cytometry is one of the diverse high-throughput methods capable of examining more than 50 features per cell. However, utilising cytometry at its full potential requires the development of optimized assays. Additionally, the resulting high-dimensional data represent a challenge for existing computational techniques. This thesis attempts to address these challenges. The first part of the thesis is focused on developing a non-linear embedding algorithm for rapid analysis of cytometry datasets called EmbedSOM. The comparison of EmbedSOM with other state-of-the-art algorithms suggested the superiority of EmbedSOM with faster runtime. This is critical for the analysis of large datasets with millions of cells. Furthermore, EmbedSOM has additional functionality such as landmark guided...

Anti-CD19 CAR-T cell immunotherapy is a hopeful treatment option for patients with B cell lymphomas, however it copes with partly severe adverse effects like neurotoxicity. Single-cell resolved molecular data sets in combination with clinical parametrization allow for comprehensive characterization of cellular subpopulations, their transcriptomic states, and their relation to the adverse effects. We here present a re-analysis of single-cell RNA sequencing data of 24 patients comprising more than 130,000 cells with focus on cellular states and their association to immune cell related neurotoxicity. For this, we developed a single-cell data portraying workflow to disentangle the transcriptional state space with single-cell resolution and its analysis in terms of modularly-composed cellular programs. We demonstrated capabilities of single-cell data portraying to disentangle transcriptional states using intuitive visualization, functional mining, molecular cell stratification, and variability analyses. Our analysis revealed that the T cell composition of the patient’s infusion product as well as the spectrum of their transcriptional states of cells derived from patients with low ICANS grade do not markedly differ from those of cells from high ICANS patients, while the relative abundancies, particularly that of cycling cells, of LAG3-mediated exhaustion and of CAR positive cells, vary. Our study provides molecular details of the transcriptomic landscape with possible impact to overcome neurotoxicity.