京都大学 / 新制・課程博士 / 博士(工学) / 甲第24146号 / 工博第5033号 / 新制||工||1786(附属図書館) / 京都大学大学院工学研究科航空宇宙工学専攻 / (主査)教授 藤本 健治, 教授 松野 文俊, 教授 森本 淳 / 学位規則第4条第1項該当 / Doctor of Philosophy (Engineering) / Kyoto University / DFAM

With the abundance of biological data, computational prediction of gene regulatory networks (GRNs) from gene expression data has become more feasible. Although incorporating other prior knowledge (PK), along with gene expression, greatly improves prediction accuracy, the accuracy remains low. PK in GRN inference can be categorized into noisy and curated. Several algorithms were proposed to incorporate noisy PK, but none address curated PK. Another challenge is that much of the PK is not stored in databases or not in a unified structured format to be accessible by inference algorithms. Moreover, no GRN inference method exists that supports post-prediction PK. This thesis addresses those limitations with three solutions: PEAK algorithm for integrating both curated and noisy PK, Online-PEAK for post-prediction interactive feedback, and DeTangle for visualization and navigation of GRNs. PEAK integrates both curated as well as noisy PK in GRN inference. We introduce a novel method for GRN inference, CurInf, to effectively integrate curated PK, and we use the previous method, Modified Elastic Net, for noisy PK, and we call it NoisInf. Using 100% curated PK, CurInf improves the AUPR accuracy score over NoisInf by 27.3% in synthetic data, 86.5% in E. coli data, and 31.1% in S. cerevisiae data. Moreover, we developed an online algorithm, online-PEAK, that enables the biologist to interact with the inference algorithm, PEAK, through a visual interface to add their domain experience about the structure of the GRN as a feedback to the system. We experimentally verified the ability of online-PEAK to achieve incremental accuracy when PK is added by the user, including true and false PK. Even when the noise in PK is 10 times more than true PK, online-PEAK performs better than inference without any PK. Finally, we present DeTangle, a Web server for interactive GRN prediction and visualization. DeTangle provides a seamless analysis of GRN starting from uploading gene expression, GRN inference, post-prediction feedback using online-PEAK, and visualization and navigation of the predicted GRN. More accurate prediction of GRN can facilitate studying complex molecular interactions, understanding diseases, and aiding drug design. / Ph. D. / Proteins are complex molecules that are responsible for most of the functionalities in living organisms. Proteins are produced from genes inside the cells. The production of proteins can be controlled by other proteins causing their production to increase or to decrease. This control is called gene regulation. Studying gene regulation between genes is important because it facilitates understanding diseases and aiding drug design. In this thesis, I developed computational methods, called PEAK and online-PEAK, to predict the interactions between genes using computer algorithms. My test results show that the method PEAK is more accurate than previous existing methods. I have also built a Web application, DeTangle, to help biologists use the algorithm and visualize the results.

The recognition of prior learning (RPL) is an assessment process through which experience gained outside academic contexts is recognised. The purpose of the research was to determine the influence the learner profile has on RPL assessment. To realise the purpose, the learner profile was described and consequently learning outcomes formulated for a portfolio workshop. The method which the study followed was a qualitative interpretative approach. The research comprised a literature study about adult learning theories that served as a theoretical framework for the study, as well as an overview of RPL assessment, followed by an empirical study. The empirical research component followed a multiple data collection method. The personal life stories of the research participants were analysed in order to compile a learner profile. The role the workshop played was determined from a learner perspective by means of an open questionnaire for participants. It included a collection of workshop photographs. Finally, the reflection of a group of RPL learners about their personal learning experiences was analysed. The research found that the learner profile included distinctive personal traits, such as motivation, task orientation, a sense of responsibility and an orientation towards the future. The nature of the learner profile as well as of informal learning gained within diverse learning contexts require directed preparation for assessment. The learners regarded the portfolio workshop as a sustainable learning process and as a transformational learning experience. The role the workshop played was multidimensional, as it served as preparation for compiling the portfolio. It also empowered and prepared the learner on a personal level for the context of tertiary education. Based on the research findings, the recommendations for an RPL approach should include the following: Firstly, the learner‟s voice should be afforded recognition. Secondly, preparation for assessment is essential. Thirdly, the preparation should follow an approach of assessment as a sustainable learning process. Finally, the learning outcomes for the portfolio workshop make a contribution to support the learners in bridging the learning contexts of informal learning and academic learning. The research makes a valuable contribution, as the empirical research shows that the holistic learner profile guides the approach to assessment. The mission of RPL to bring about transformation will only become a reality if a learner-centred approach recognises and empowers the learner on a personal and academic level. / Die Erkenning van Vorige Leer (EVL) is 'n assesseringsproses waardeur ervaring wat buite akademiese kontekste verwerf is, erken word. Die doel met die navorsing was om die invloed van die leerderprofiel op EVL-assessering te bepaal. Om die doel te bereik, is die leerderprofiel beskryf en voortvloeiend daaruit leeruitkomste vir 'n portefeuljewerkswinkel saamgestel. Die metode wat die studie gevolg het, was 'n kwalitatiewe interpretiewe benadering. Die navorsing bestaan uit 'n literatuurstudie oor volwasse leerteorieë wat as teoretiese raamwerk vir die studie gedien het, asook 'n oorsig oor EVL-assessering, gevolg deur 'n empiriese studie. Die empiriese navorsingskomponent het 'n meervoudige dataversamelingsmetode gevolg. Die navorsingsdeelnemers se persoonlike lewensverhale is geanaliseer om sodoende 'n leerderprofiel saam te stel. Die rol van die werkswinkel is vanuit 'n leerdersperspektief bepaal deur middel van 'n ope-vraelys aan deelnemers, wat 'n versameling werkswinkel-foto's ingesluit het. Laastens is die reflektering van 'n groepie EVL-leerders oor hulle persoonlike leerervaringe ge-analiseer. Die navorsing het bevind dat die leerderprofiel onderskeidende persoonlike kenmerke, soos motivering, taakgerigtheid, 'n verantwoodelikheidsin en 'n toekomsgerigtheid insluit. Die aard van die leerderprofiel, asook die aard van informele leer verwerf binne diverse leerkontekste vereis gerigte voorbereiding tot assessering. Die leerders het die portefeuljewerkswinkel as 'n volhoubare leerproses en as transformasionele leerervaring beskou. Die rol van die werkswinkel is meervoudig, aangesien dit dien as voorbereiding tot die saamstel van die portefeulje, asook die leerder op persoonlike vlak bemagtig en voorberei vir die konteks van tersiêre onderrig. Die aanbevelings tot 'n EVL-benadering behoort, op grond van die navorsingsbevindinge, die volgende in te sluit. Eerstens moet erkenning aan die stem van die leerder gegee word, tweedens is voorbereiding tot assessering noodsaaklik en derdens behoort die voorbereiding 'n benadering van assessering as 'n volhoubare leerposes te volg. Laastens lewer die leeruitkomste vir die portefeuljewerkswinkel 'n praktykgerigte bydrae om die leerder te ondersteun om die leerkontekste van informele leer en akademiese leer te oorbrug. Die navorsing lewer 'n waardevolle bydrae aangesien die empiriese navorsing toon dat die holistiese leerderprofiel die benadering tot assessering rig. Die missie van EVL om transformasie te weeg te bring, sal slegs 'n realiteit word indien 'n leerdergesentreerde benadering die leerder op persoonlike en op akademiese vlak erken en bemagtig. / Curriculum and Instructional Studies / D. Ed. (Curriculum Studies)

La reconnaissance des acquis de l'expérience fait référence à des mesures permettant d'identifier et de valoriser les acquis des apprentissages réalisés en dehors des établissements d'enseignement et de formation. À présent, les dispositions de ce type existent ou se mettent en place dans de nombreux pays, et elles sont également promues dans le cadre de la coopération européenne en éducation et formation. Toutefois, la reconnaissance des acquis de l'expérience ne se réalise pas nécessairement dans tous les secteurs de l'éducation et de la formation, sa mise en place étant particulièrement délicate dans le secteur de l'enseignement supérieur. Partant de ces éléments, cette recherche se penche sur trois systèmes d'enseignement supérieur relativement avancés en la matière : l'Angleterre, la Belgique francophone et la France. En interrogeant l'institutionnalisation et la mise en oeuvre de la reconnaissance des acquis de l'expérience dans ces trois contextes, la recherche pose la question de savoir si, sous l'influence des processus supranationaux, les dispositifs de reconnaissance des acquis de l'expérience convergent vers les mêmes modalités de réalisation. Une triple perspective est adoptée pour explorer cette question. Premièrement, il s'agit d'examiner le rôle de la coopération européenne dans la mise en place des dispositifs nationaux de reconnaissance des acquis de l'expérience. Deuxièmement, il s'agit d'étudier le cheminement de la reconnaissance des acquis de l'expérience dans les trois systèmes investigués. Troisièmement, il s'agit de mettre en lumière la façon dont se réalise la reconnaissance des acquis de l'expérience dans les établissements d'enseignement supérieur, en l'occurrence les universités. Les deux premiers volets de la recherche s'appuient principalement sur le recueil et l'analyse des sources écrites, y compris la réglementation, les lignes directrices officielles, les rapports d'analyse et les travaux de recherche. L'investigation au niveau institutionnel est centrée sur un échantillon de trois à quatre universités dans chaque système étudié, et elle s'appuie sur des visites de sites, des entretiens avec des acteurs impliqués dans la reconnaissance des acquis de l'expérience et l'information diffusée par les universités. La recherche met en évidence le fait que les dispositifs de reconnaissance des acquis de l'expérience s'élaborent en fonction de nombreux paramètres tels que l'organisation et la structuration de l'enseignement supérieur, la perception du rôle de l'enseignement supérieur ou encore la présence des acteurs individuels et collectifs aptes à promouvoir ces dispositifs. La mouvance européenne autour de la reconnaissance des acquis de l'expérience apparaît ainsi comme un cadre de référence général, prêt à se décliner différemment dans chaque système d'enseignement supérieur. / The recognition of prior experiential learning refers to arrangements that aim to identify and valorise learning outcomes achieved outside education and training institutions. At present, such arrangements exist or are being introduced in a number of countries, and are also promoted in the framework of European cooperation in education and training. However, the recognition of prior experiential learning does not always apply to all sectors of education and training, its implementation being particularly challenging in the higher education sector. Starting from these elements, this research looks at three higher education systems considered relatively advanced in this field: England, French-speaking Belgium and France. By researching the institutionalisation and implementation of the recognition of prior experiential learning in the aforementioned systems, the study raises the question of whether, under the influence of supranational processes, approaches to the recognition of prior experiential learning converge towards the same model. A three-point perspective is adopted to explore this issue. First, the research examines the role of European cooperation in the development of national arrangements for the recognition of prior experiential learning. Second, the study looks at the evolution of the recognition of prior experiential learning in the investigated systems. Finally, the analysis considers how the recognition of prior experiential learning is being implemented in higher education institutions, in this case, universities. The investigation related to the first two areas mainly relies on the collection and analysis of written sources, including regulations, official guidelines, analysis reports and research. The institutional analysis focuses on a sample of three to four universities in each studied system, and is based on site visits, interviews with actors involved in the recognition of prior experiential learning and the information disseminated by universities. Research shows that the development of arrangements for the recognition of prior experiential learning is influenced by a number of factors, including the organisation and structure of higher education systems, the perception of the role of higher education, as well as the presence of individual and collective actors able to promote these arrangements. Thus, European activities in the area of the recognition of prior experiential learning appear as a general frame of reference that translates into different arrangements in each higher education system.

Dans cette thèse nous nous intéressons aux problèmes de la sélection de variables en régression linéaire. Ces travaux sont en particulier motivés par les développements récents en génomique, protéomique, imagerie biomédicale, traitement de signal, traitement d’image, en marketing, etc… Nous regardons ce problème selon les deux points de vue fréquentielle et bayésienne.Dans un cadre fréquentiel, nous proposons des méthodes pour faire face au problème de la sélection de variables, dans des situations pour lesquelles le nombre de variables peut être beaucoup plus grand que la taille de l’échantillon, avec présence possible d’une structure supplémentaire entre les variables, telle qu’une forte corrélation ou un certain ordre entre les variables successives. Les performances théoriques sont explorées ; nous montrons que sous certaines conditions de régularité, les méthodes proposées possèdent de bonnes propriétés statistiques, telles que des inégalités de parcimonie, la consistance au niveau de la sélection de variables et la normalité asymptotique.Dans un cadre bayésien, nous proposons une approche globale de la sélection de variables en régression construite sur les lois à priori g de Zellner dans une approche similaire mais non identique à celle de Liang et al. (2008) Notre choix ne nécessite aucune calibration. Nous comparons les approches de régularisation bayésienne et fréquentielle dans un contexte peu informatif où le nombre de variables est presque égal à la taille de l’échantillon. / We are interested in variable sélection in linear régression models. This research is motivated by recent development in microarrays, proteomics, brain images, among others. We study this problem in both frequentist and bayesian viewpoints.In a frequentist framework, we propose methods to deal with the problem of variable sélection, when the number of variables is much larger than the sample size with a possibly présence of additional structure in the predictor variables, such as high corrélations or order between successive variables. The performance of the proposed methods is theoretically investigated ; we prove that, under regularity conditions, the proposed estimators possess statistical good properties, such as Sparsity Oracle Inequalities, variable sélection consistency and asymptotic normality.In a Bayesian Framework, we propose a global noninformative approach for Bayesian variable sélection. In this thesis, we pay spécial attention to two calibration-free hierarchical Zellner’s g-priors. The first one is the Jeffreys prior which is not location invariant. A second one avoids this problem by only considering models with at least one variable in the model. The practical performance of the proposed methods is illustrated through numerical experiments on simulated and real world datasets, with a comparison betwenn Bayesian and frequentist approaches under a low informative constraint when the number of variables is almost equal to the number of observations.

Through prior learning assessment (PLA) programs, students' professional and life experiences are evaluated to determine if they equate to college program courses-removing the need for the student to complete traditional college courses to earn a postsecondary degree. The purpose of this study was to examine the perspectives of faculty at the study site on the academic preparedness of nontraditional students. Knowles' andragogy theory guided the research questions on the academic preparedness of students earning prior learning credits and what characteristics of PLA programs faculty believe contribute to the academic success of these students. Thirty-six participants completed a survey where they provided narratives that were coded to determine overarching themes. The findings of this case study suggested that faculty believed that nontraditional students who earned credits through PLA are independent, motivated, and knowledgeable, but whether or not they are academically prepared seemed to vary by academic program. The findings also suggested that, while PLA programs have limitations because of student perceptions that may conflict with reality and the limited usability of PLA across academic programs, PLA programs provide a time benefit to students by allowing a shorter time to degree. The study supports positive social change by providing a resource for future development and growth of PLA programs to expand student access and options in obtaining postsecondary credentials while also lowering postsecondary education cost.

The recognition of prior learning (RPL) is an assessment process through which experience gained outside academic contexts is recognised. The purpose of the research was to determine the influence the learner profile has on RPL assessment. To realise the purpose, the learner profile was described and consequently learning outcomes formulated for a portfolio workshop. The method which the study followed was a qualitative interpretative approach. The research comprised a literature study about adult learning theories that served as a theoretical framework for the study, as well as an overview of RPL assessment, followed by an empirical study. The empirical research component followed a multiple data collection method. The personal life stories of the research participants were analysed in order to compile a learner profile. The role the workshop played was determined from a learner perspective by means of an open questionnaire for participants. It included a collection of workshop photographs. Finally, the reflection of a group of RPL learners about their personal learning experiences was analysed. The research found that the learner profile included distinctive personal traits, such as motivation, task orientation, a sense of responsibility and an orientation towards the future. The nature of the learner profile as well as of informal learning gained within diverse learning contexts require directed preparation for assessment. The learners regarded the portfolio workshop as a sustainable learning process and as a transformational learning experience. The role the workshop played was multidimensional, as it served as preparation for compiling the portfolio. It also empowered and prepared the learner on a personal level for the context of tertiary education. Based on the research findings, the recommendations for an RPL approach should include the following: Firstly, the learner‟s voice should be afforded recognition. Secondly, preparation for assessment is essential. Thirdly, the preparation should follow an approach of assessment as a sustainable learning process. Finally, the learning outcomes for the portfolio workshop make a contribution to support the learners in bridging the learning contexts of informal learning and academic learning. The research makes a valuable contribution, as the empirical research shows that the holistic learner profile guides the approach to assessment. The mission of RPL to bring about transformation will only become a reality if a learner-centred approach recognises and empowers the learner on a personal and academic level. / Die Erkenning van Vorige Leer (EVL) is 'n assesseringsproses waardeur ervaring wat buite akademiese kontekste verwerf is, erken word. Die doel met die navorsing was om die invloed van die leerderprofiel op EVL-assessering te bepaal. Om die doel te bereik, is die leerderprofiel beskryf en voortvloeiend daaruit leeruitkomste vir 'n portefeuljewerkswinkel saamgestel. Die metode wat die studie gevolg het, was 'n kwalitatiewe interpretiewe benadering. Die navorsing bestaan uit 'n literatuurstudie oor volwasse leerteorieë wat as teoretiese raamwerk vir die studie gedien het, asook 'n oorsig oor EVL-assessering, gevolg deur 'n empiriese studie. Die empiriese navorsingskomponent het 'n meervoudige dataversamelingsmetode gevolg. Die navorsingsdeelnemers se persoonlike lewensverhale is geanaliseer om sodoende 'n leerderprofiel saam te stel. Die rol van die werkswinkel is vanuit 'n leerdersperspektief bepaal deur middel van 'n ope-vraelys aan deelnemers, wat 'n versameling werkswinkel-foto's ingesluit het. Laastens is die reflektering van 'n groepie EVL-leerders oor hulle persoonlike leerervaringe ge-analiseer. Die navorsing het bevind dat die leerderprofiel onderskeidende persoonlike kenmerke, soos motivering, taakgerigtheid, 'n verantwoodelikheidsin en 'n toekomsgerigtheid insluit. Die aard van die leerderprofiel, asook die aard van informele leer verwerf binne diverse leerkontekste vereis gerigte voorbereiding tot assessering. Die leerders het die portefeuljewerkswinkel as 'n volhoubare leerproses en as transformasionele leerervaring beskou. Die rol van die werkswinkel is meervoudig, aangesien dit dien as voorbereiding tot die saamstel van die portefeulje, asook die leerder op persoonlike vlak bemagtig en voorberei vir die konteks van tersiêre onderrig. Die aanbevelings tot 'n EVL-benadering behoort, op grond van die navorsingsbevindinge, die volgende in te sluit. Eerstens moet erkenning aan die stem van die leerder gegee word, tweedens is voorbereiding tot assessering noodsaaklik en derdens behoort die voorbereiding 'n benadering van assessering as 'n volhoubare leerposes te volg. Laastens lewer die leeruitkomste vir die portefeuljewerkswinkel 'n praktykgerigte bydrae om die leerder te ondersteun om die leerkontekste van informele leer en akademiese leer te oorbrug. Die navorsing lewer 'n waardevolle bydrae aangesien die empiriese navorsing toon dat die holistiese leerderprofiel die benadering tot assessering rig. Die missie van EVL om transformasie te weeg te bring, sal slegs 'n realiteit word indien 'n leerdergesentreerde benadering die leerder op persoonlike en op akademiese vlak erken en bemagtig. / Curriculum and Instructional Studies / D. Ed. (Curriculum Studies)

Gravitational lensing, a phenomenon in astronomy, occurs when the gravitational field of a massive object, such as a galaxy or a black hole, bends the path of light from a distant object behind it. This bending results in a distortion or magnification of the distant object's image, often seen as arcs or rings surrounding the foreground object. The Starlet wavelet transform offers a robust approach to representing galaxy images sparsely. This technique breaks down an image into wavelet coefficients at various scales and orientations, effectively capturing both large-scale structures and fine details. The Starlet wavelet transform offers a robust approach to representing galaxy images sparsely. This technique breaks down an image into wavelet coefficients at various scales and orientations, effectively capturing both large-scale structures and fine details. The horseshoe prior has emerged as a highly effective Bayesian technique for promoting sparsity and regularization in statistical modeling. It aggressively shrinks negligible values while preserving important features, making it particularly useful in situations where the reconstruction of an original image from limited noisy observations is inherently challenging. The main objective of this thesis is to apply sparse regularization techniques, particularly the horseshoe prior, to reconstruct the background source galaxy from gravitationally lensed images. By demonstrating the effectiveness of the horseshoe prior in this context, this thesis tackles the challenging inverse problem of reconstructing lensed galaxy images. Our proposed methodology involves applying the horseshoe prior to the wavelet coefficients of lensed galaxy images. By exploiting the sparsity of the wavelet representation and the noise-suppressing behavior of the horseshoe prior, we achieve well-regularized reconstructions that reduce noise and artifacts while preserving structural details. Experiments conducted on simulated lensed galaxy images demonstrate lower mean squared error and higher structural similarity with the horseshoe prior compared to alternative methods, validating its efficacy as an efficient sparse modeling technique. / Les lentilles gravitationnelles se produisent lorsque le champ gravitationnel d'un objet massif dévie la trajectoire de la lumière provenant d'un objet lointain, entraînant une distorsion ou une amplification de l'image de l'objet lointain. La transformation Starlet fournit une méthode robuste pour obtenir une représentation éparse des images de galaxies, capturant efficacement leurs caractéristiques essentielles avec un minimum de données. Cette représentation réduit les besoins de stockage et de calcul, et facilite des tâches telles que le débruitage, la compression et l'extraction de caractéristiques. La distribution a priori de fer à cheval est une technique bayésienne efficace pour promouvoir la sparsité et la régularisation dans la modélisation statistique. Elle réduit de manière agressive les valeurs négligeables tout en préservant les caractéristiques importantes, ce qui la rend particulièrement utile dans les situations où la reconstruction d'une image originale à partir d'observations bruitées est difficile. Étant donné la nature mal posée de la reconstruction des images de galaxies à partir de données bruitées, l'utilisation de la distribution a priori devient cruciale pour résoudre les ambiguïtés. Les techniques utilisant une distribution a priori favorisant la sparsité ont été efficaces pour relever des défis similaires dans divers domaines. L'objectif principal de cette thèse est d'appliquer des techniques de régularisation favorisant la sparsité, en particulier la distribution a priori de fer à cheval, pour reconstruire les galaxies d'arrière-plan à partir d'images de lentilles gravitationnelles. Notre méthodologie proposée consiste à appliquer la distribution a priori de fer à cheval aux coefficients d'ondelettes des images de galaxies lentillées. En exploitant la sparsité de la représentation en ondelettes et le comportement de suppression du bruit de la distribution a priori de fer à cheval, nous obtenons des reconstructions bien régularisées qui réduisent le bruit et les artefacts tout en préservant les détails structurels. Des expériences menées sur des images simulées de galaxies lentillées montrent une erreur quadratique moyenne inférieure et une similarité structurelle plus élevée avec la distribution a priori de fer à cheval par rapport à d'autres méthodes, validant son efficacité.

No description available.

In many areas of the agriculture, biological, physical and social sciences, spatial lattice data are becoming increasingly common. In addition, a large amount of lattice data shows not only visible spatial pattern but also temporal pattern (see, Zhu et al. 2005). An interesting problem is to develop a model to systematically model the relationship between the response variable and possible explanatory variable, while accounting for space and time effect simultaneously. Spatial-temporal linear model and the corresponding likelihood-based statistical inference are important tools for the analysis of spatial-temporal lattice data. We propose a general asymptotic framework for spatial-temporal linear models and investigate the property of maximum likelihood estimates under such framework. Mild regularity conditions on the spatial-temporal weight matrices will be put in order to derive the asymptotic properties (consistency and asymptotic normality) of maximum likelihood estimates. A simulation study is conducted to examine the finite-sample properties of the maximum likelihood estimates. For spatial data, aside from traditional likelihood-based method, a variety of literature has discussed Bayesian approach to estimate the correlation (auto-covariance function) among spatial data, especially Zheng et al. (2010) proposed a nonparametric Bayesian approach to estimate a spectral density. We will also discuss nonparametric Bayesian approach in analyzing spatial data. We will propose a general procedure for constructing a multivariate Feller prior and establish its theoretical property as a nonparametric prior. A blocked Gibbs sampling algorithm is also proposed for computation since the posterior distribution is analytically manageable.