Tensors are of great interest to many applications in engineering and in medical imaging, but a proper analysis and visualization remains challenging. Physics-based visualization of tensor fields has proven to show the main features of symmetric second-order tensor fields, while still displaying the most important information of the data, namely the main directions in medical diffusion tensor data using texture and additional attributes using color-coding, in a continuous representation. Nevertheless, its application and usability remains limited due to its computational expensive and sensitive nature. We introduce a novel approach to compute a fabric-like texture pattern from tensor fields motivated by image-space line integral convolution (LIC). Although, our approach can be applied to arbitrary, non-selfintersecting surfaces, we are focusing on special surfaces following neural fibers in the brain.We employ a multipass rendering approach whose main focus lies on regaining three-dimensionality of the data under user interaction as well as being able to have a seamless transition between local and global structures including a proper visualization of degenerated points.

The East African Rift System diverges in the Lake Natron Basin of Northern Tanzania and is a major zone of continental extension and crustal thinning with resulting in active tectonics and volcanism. The discovery of Acheulean technology in Olduvai Gorge and Peninj as well as the presence of significant volcanic centers, has made in the region subject to studies in various disciplines. However, lack of precise radiometric age constraints due to the complex geology of the region is a major drawback. The basin is bordered on the western side by an escarpment that contains thick sequences of volcanic (nephelinites, basanites, hawaiites, alkali basalts), volcaniclastic and lacustrine strata that predates 1.2 Ma. This thesis is based on 41 rock samples that were collected from two geological sections, the Endukai Kete (EK) and Waterfall (WF) sections and aims to establish a preliminary geomagnetic polarity time scale (GPTS) for the Natron Escarpment, together with establishing possible flow directions of the volcanic lavas within these sections. Nephelinites of EK section have an inferred NW-SE direction of flow, based on study of anisotropy of magnetic susceptibility. They record a normal polarity that most likely correspond to the Cobb Mountain Event (CMT; 1.187-1.208 Ma), although there is an 80-ka discrepancy between the CMT event and the dated lavas. The most probable source is the Mosonik that erupted nephelinitic lavas 1.28 Ma ago. The palagonitic tuff layer below the nephelinites displays reverse polarity and a NE-SW direction of flow. Due to the absence of approximately 200 m strata within the basanite series of the section, regional lithological correlation is used to constrain the GPTS pattern. Hajaro Beds of the Peninj Group to the north of the escarpment, postdates the Olduvai Event (1.71 to 1.86 Ma) and lacustrine strata of the escarpment for EK and WF sections are deposited over the same unconformity and share depositional similarities. Therefore, the lacustrine strata are correlative to Hajaro beds and the normal event observed within the basanite series of both sections is attributed to the Réunion Event (2.116 – 2.137 Ma). The establishment of a preliminary magnetostratigraphic sequence presented in this thesis demonstrate that the rift escarpment in northern Tanzania is suitable for paleomagnetic dating. Future studies should be conducted to establish a more detailed and constrained magnetostratigraphic section, which will be of great use in this part of the African Rift where radiometric dating has been challenging.

Orientador: Nilton Nobuhiro Imai / Resumo: O uso do UAV no sensoriamento remoto é uma área crescente de conhecimento e pressionou o desenvolvimento de câmeras multi e hiperespectrais leves, que poderiam ser usadas embarcadas em UAV. Como as informações espectrais dependem das condições de iluminação da cena, as imagens adquiridas por UAV exigem pesquisa para avaliar o processamento de imagens para melhor adaptar os conceitos já estabelecidos às imagens orbitais. Portanto, a correção radiométrica é de fundamental importância para a extração de dados de imagens com alta confiança, uma vez que se sabe que o fator de refletância é uma função da estrutura geométrica, ângulo solar e propriedades ópticas. Nesse sentido, foram desenvolvidas metodologias para corrigir imagens das diferenças de iluminação, utilizando as Funções de Distribuição de Refletância Bidirecional e os Modelos de Correção Topográfica, também conhecidos como correção de iluminação. Este trabalho utiliza a câmera hiperespectral Rikola a bordo de um UAV em latitudes tropicais. Ele avalia como a anisotropia pode influenciar a variabilidade na reflectância dos alvos e como os modelos de correção topográfica podem ser aplicados, usando a micro topografia, para atenuar esses efeitos. Três testes foram realizados para estudar i) as geometrias de visada da câmera hiperespectral Rikola e a disponibilidade de dados fora do Nadir, ii) a variação do fator de anisotropia entre os alvos nas geometrias de visada e iii) modelos de correção de microtopografia para corrigi... (Resumo completo, clicar acesso eletrônico abaixo) / Abstract: Use of UAV in remote sensing is a growing area of knowledge and pressed the development of lightweight multi and hyperspectral cameras, which could be used on board of UAV. Since the spectral information depends on the lighting condition of the scene, UAV acquired images demands research to assess image processing to better adapt the concepts already stablished to orbital images. Therefore, the radiometric correction is of main importance for data extraction from imagery with high confidence, once it is known that the reflectance factor is a function of geometric structure, solar angle and optical properties. In this regard, methodologies were developed to correct images from illumination differences, using the Bidirectional Reflectance Distribution Functions and the Topographic Correction Models, also known as illumination correction. This Work uses Rikola hyperspectral camera onboard of a UAV in tropical latitudes. It assesses how the anisotropy can influence variability in reflectance of targets, and how topographic correction models can be applied, using micro topography, to attenuate these effects. Three tests were done to study i) the view geometries of Rikola Hyperspectral camera and the no-Nadir data availability, ii) the variation of Anisotropy Factor between targets in the many view geometries and iii) Microtopography correction models to correct illumination differences using a highly detailed DSM (10 cm and 3 cm) to assess the micro relief. We applied the correcti... (Complete abstract click electronic access below) / Mestre

Experimental characterization of the mechanical properties in a thin injection moulded Low-Density Polyethylene (LDPE) plate is per- formed in this work. Anisotropy in LDPE at different material orientations is measured from the Digital Image Correlation (DIC) observation of the specimens during uniaxial tensile test. From the test response and observation from DIC, the studied material is found to be significantly anisotropic. Finite Element simulation (FE-simulation) of in-plane anisotropy of material is carried in AbaqusTM R2020 using available models like Hill48 and Barlat2004. When necessary the simulation plastic potentials for these models are optimized against experimental yield stress ratio (R) and anisotropic ratio (r). To express the nonlinear mechanical behavior, a suitable hardening extrapolation model, namely Swift/Hockett-Sherby is selected from several extrapolation models based on experimental data. To validate the experimental methods, simulation methods and material characterization process, finite element simulation results such as force displacement, strain distribution and different anisotropic related properties are compared with the experimental data. Finally, advantages and disadvantages of different simulation models are discussed.

Charles University Faculty of Pharmacy in Hradec Králové Department of Pharmacology and Toxicology Student: Hana Danková Supervisors: Prof. Ago Rinken, PhD. MSc. Tõnis Laasfeld PharmDr. Ivan Vokřál, PhD. Title of diploma thesis: Characetrization of ligand binding to M1 muscarinic receptor using fluorescence anisotropy method Muscarinic acetylcholine receptors (mAChRs), members of the superfamily of G-protein coupled receptors (GPCRs), regulate vital physiological processes and are important targets in drug research. Five different subtypes (M1 - M5) have been identified. M1 mAChR is mainly distributed in the central nervous system and is linked to pathophysiology of neurodegenerative diseases. In recent years, fluorescent methods have been frequently used in studies of ligand binding to receptors. The fluorescence anisotropy (FA) is a homogenous assay to characterize ligand binding to receptors. In this work, we have evaluated the FA method with fluorescent ligand MK342 binding to M1 mAChRs expressed on budded baculovirus (BBV) particles. The fluorescence ligand was binding with the high affinity (4,4 nM) to M1 receptor in constructed BBV preparation. The apparent binding affinities (pKi) of eleven classical and three bitopic muscarinic ligands were screened and compared to previously published...

Thesis: M.S., Massachusetts Institute of Technology, Department of Electrical Engineering and Computer Science, 1980 / Includes bibliographcial references. / by Farhad Hakimi. / M.S. / M.S. Massachusetts Institute of Technology, Department of Electrical Engineering and Computer Science

In this thesis, we examine the anisotropy phenomenon in popular masked language models, BERT and RoBERTa, in detail. We propose a possible explanation for this unreasonable phenomenon. First, we demonstrate that the contextualized word vectors derived from pretrained masked language model-based encoders share a common, perhaps undesirable pattern across layers. Namely, we find cases of persistent outlier neurons within BERT and RoBERTa's hidden state vectors that consistently bear the smallest or largest values in said vectors. In an attempt to investigate the source of this information, we introduce a neuron-level analysis method, which reveals that the outliers are closely related to information captured by positional embeddings. Second, we find that a simple normalization method, whitening can make the vector space isotropic. Lastly, we demonstrate that ''clipping'' the outliers or whitening can more accurately distinguish word senses, as well as lead to better sentence embeddings when mean pooling.

Philosophiae Doctor - PhD / Low- and medium-spin bands of 187Os have been studied using the AFRODITE array, following the 186W(4He,3n)187Os reaction at a beam energy of 37 MeV. The measurements of γ − γ coincidences, angular distribution ratios (RAD), polarization and γ-intensities were performed using eleven High Purity Germanium (HPGe) clover detectors. In the current work, all the previously known bands have been significantly extended and five new bands have been added to the level scheme. The observed bands are interpreted within the cranked shell model (CSM), cranked Nilsson-StrutinskyBogoliubov (CNSB) formalism and Quasiparticle-plus-Triaxial-Rotor (QTR) model. Systematic comparison of bands with the neighbouring isotopes has also been made. Comparison of the models with experimental data shows good agreement. The configurations of some of the previously observed bands have been modified. Most importantly, the coupling of 2+ γ band to the 11/2+[615] neutron configuration is observed for the first time.

In this thesis the influence of different growth conditions on the structural and the magnetic properties of magnetite were analyzed. Therefore, ultrathin Fe3O4 films were grown on MgO(001) substrates, on NiO, and on Fe pre-covered MgO(001) substrates. In the first part of this thesis magnetite films with different film thicknesses were deposited directly on MgO by RMBE to investigate the thickness dependence of the anomalous strain and the in-plane magnetic anisotropy. Surface sensitive methods like XPS and LEED have shown that all films in the investigated thickness range are stoichiometric and epitactic magnetite. Bulk sensitive XRD experiments at the specular rod point to well-ordered films with homogenous film thickness indicated by the distinct Laue oscillations. However, the vertical layer distances are smaller than expected even for strained magnetite. Raman measurements were carried out to clarify this contradiction between surface sensitive and bulk sensitive measurements. While the 20 nm and 30 nm films exhibit the typical bands for magnetite, no distinct bands can be observed for the 7.6 nm film. Due to this results we first assume a partial formation of a thin maghemite layer on top of the uncapped magnetite film under ambient conditions. Nevertheless, additional XPS measurement exclude the complete oxidation of magnetite to maghemite since there is no significantly increased Fe3+-signal visible. Thus, the low vertical layer distance can be attributed to the presence of APBs causing an anomalous strain relaxation as reported in literature. Although all films feature ferromagnetic behavior there are differences in the characteristic of the magnetic in-plane anisotropy. The 7.6 nm film has an in-plane magnetic isotropy while the 20 nm and 30 nm film have an in-plane fourfold magnetic anisotropy. Here, the fourfold magnetic anisotropy is stronger for the 20 nm magnetite film than for the 30 nm film. The critical film thickness for the transition from magnetic isotropy to magnetic fourfold anisotropy may be influenced by film thickness and lattice strain induced by the substrate. The second part of this thesis features the thickness dependence of the structural quality of Fe3O4/NiO bilayers. Each film of the Fe3O4/NiO bilayer on MgO(001) have been successfully grown by RMBE. LEED and XPS experiments have proven that the surface near regions of the distinct films have high structural and stoichiometric properties. Here, too, the detailed ’bulk’ structural characterization of Fe3O4/NiO bilayers were carried out using XRD. It was shown that the Fe3O4 films grow homogeneously and smoothly on NiO films if the NiO film thickness is below 24 nm. Above this NiO film thickness the structural quality of the magnetite films gets distinctly worse. This behavior can be attributed to the fact that the interface roughness between NiO and Fe3O4 depends on the NiO film thickness. The roughness of the 3 nm NiO film is rather small and it is rising obviously with increasing NiO film thickness. Thus, the structural quality of the magnetite films grown on 30 nm NiO films is constantly reduced with increasing magnetite film thickness since the quality of the Fe3O4 films is influenced by the quality of the Fe3O4/NiO interface. In the third study the influence of initial iron buffer layers on the magnetic properties of magnetite grown on MgO(001) substrates has been investigated. In situ XPS and LEED indicate that the structural and stoichiometric properties of the surface near region of the magnetite films are not influenced by the iron buffer layer. However, the structural and magnetic properties of the whole film have changed compared to magnetite grown directly on MgO as shown by XRD and MOKE. The crystalline quality is poor and the magnetic easy axis of the magnetic in-plane anisotropy is rotated by 45◦ compared to magnetite films grown directly on MgO. Both crystalline quality and the rotation of the magnetic in-plane anisotropy show no dependence on the film thickness. However, the strength of the magnetocrystalline anisotropy decreases with increasing film thickness. XPS and XRD measurements have indicated that the iron buffer layer is completely oxidized during the second growth stage of the magnetite. The small Kerr rotation in the MOKE experiments of the samples with film thicknesses up to 28 nm confirms this result since a remaining iron film would cause a higher Kerr rotation. In the last part of this thesis the structural and magnetic properties of a partially oxidized, a completely oxidized and a metallic iron film are analyzed. The partially oxidized iron film is a bilayer with a metallic iron film and an iron oxide film. The surface near stoichiometry of both oxidized iron films correspond to magnetite as proven by XPS. The structural analysis by XRD reveals that though these magnetite films are crystalline, they have an inhomogeneous thickness. The magnetization curves of the partially oxidized film (bilayer) measured by MOKE exhibit a magnetic saturation which is comparable to the magnetic saturation of the single metallic iron film. However, the coercive field is higher compared to the single metallic iron film due to the high interface roughnesses. Considering the coercive field as a function of the sample rotation α the bilayer exhibits a fourfold anisotropy with eight maxima. However, the angular dependence of the magnetic remanence features a simple fourfold anisotropy with easy axes in h110i directions of MgO(001). Vector MOKE analysis displays that the magnetic reversal processes of the bilayer are similar to single metallic iron films. Thus, the bilayer exhibits mostly the magnetic properties of a single iron film. The bilayer has the same magnetic easy axis and a similar magnetic saturation. The magnetic easy axis of the magnetic in-plane anisotropy of the completely oxidized iron film is rotated by 45◦ compared to magnetite films grown directly on MgO as already observed in Ref. [148]. The completely oxidized iron film exhibits also a significantly increased coercive field due to high surface roughness. A magneto-dynamic investigation of the exchange-coupling of the Fe3O4/Fe bilayer system was carried out to yield all relevant magnetic parameters, such as anisotropies, as well as the coupling constant J1. Here, also a complex fourfold anisotropy was observed, which might be due to a coupling of a perpendicular spin-wave mode in the magnetite layer with the acoustical coupling mode. We have also successfully calculated the angular dependence of the resonance field of all films using the eighth order of the magnetocrystalline anisotropy. Although we have applied many measurement methods, we have found no explanation for the complex fourfold angular dependence of the coercive field of the bilayer. Nevertheless, this bilayer provides interesting properties for application in MTJs due to its enhanced magnetic properties like complex fourfold magnetic anisotropy and higher coercive field and remanence. All in all, we have shown that the structural and magnetic properties of magnetite films are strongly influenced by interlayers between film and substrate. While magnetite films directly deposited on MgO exhibit a homogeneous film thickness, both NiO interlayers with a thickness above 24 nm and initially grown iron films deteriorate the structural quality of the on top grown magnetite films. In addition, the magnetic fourfold anisotropy is rotated by 45◦ in comparison to magnetite films grown directly on MgO for the structurally disturbed magnetite on iron pre-covered MgO substrates. Here, further investigations are necessary to understand why this anisotropy rotation occurs and how we can improve the structural quality of magnetite on iron and NiO. The growth of magnetite by oxidizing previously deposited iron films leads to crystalline magnetite films but with inhomogeneous film thicknesses. In case of a Fe3O4/Fe bilayer, the magnetic properties are drastically changed due to the magnetic coupling between the iron and the magnetite film.

Une cellule est soumise à différents stimuli internes et externes. Pour remplir ses fonctions, elle doit donc s’adapter rapidement. Cela implique une régulation fine de l’expression génique. Celle-ci s’effectue au niveau transcriptionnel, mais également au niveau traductionnel. La traduction comprend trois phases : le démarrage, l’allongement et la terminaison. C’est au cours du démarrage de la traduction que s’effectue la sélection du codon de démarrage et donc le choix du cadre de lecture de l’ARNm. D’un point de vue cinétique, le démarrage de la traduction est l’étape limitante. Ainsi, il apparait comme une cible privilégiée pour le contrôle traductionnel.Chez les archées, le démarrage de la traduction met en jeu un complexe macromoléculaire formé de la petite sous-unité du ribosome, d’un ARNm, d’un ARN de transfert initiateur méthionylé (Met-ARNtiMet) et de trois facteurs de démarrage de la traduction (aIF1, aIF1A et aIF2). De manière intéressante, ces trois facteurs de démarrage ont chacun un orthologue eucaryote.Les ARNti archées et eucaryotes possèdent une paire de bases très conservée A1-U72, au sommet de la tige acceptrice. Cette paire de base a été montrée importante pour la discrimination des ARNt initiateurs et élongateurs. De plus, des travaux suggèrent l’importance de la géométrie de la paire A1-U72 pour l’identité initiatrice de ces ARNts. Cependant, au début de ma thèse, aucune donnée structurale n’était disponible pour expliquer comment les caractéristiques de la paire A1-U72 participaient à la sélection de l’ARNt initiateur. Dans un premier temps, mon travail de thèse a consisté en la construction d’une souche bactérienne d’E.coli utilisant comme seul source d’ARNti un variant d’ARNt initiateur bactérien (ARNtfMet) possédant une paire de base A1-U72 (ARNtfMetA1-U72). L’utilisation de cette souche nous a permis d’obtenir de grandes quantités d’ARNtfMetA1-U72 purifié. De plus, la structure cristallographique de cet ARNtA1-U72 a pu être déterminée à 2.8 Å de résolution. Un arrangement inhabituel des bases A1 et U72 a été observé.Tous les acteurs du démarrage de la traduction de l’archée P. abyssi étant disponibles au laboratoire, une étude du complexe de démarrage de la traduction archée par cryo-microscopie électronique a pu être réalisée. L’étude a permis d’identifier deux conformations de l’ARNti dans le complexe de démarrage, IC0-Premote (5.3 Å de résolution) et IC1-Pin (7.5 Å de résolution). Ces deux conformations permettent de proposer un modèle pour l’accommodation de l’ARNt initiateur lors de l’appariement au codon de démarrage.Finalement, je me suis également intéressée au rôle du facteur aIF1. La disponibilité de structures 3D et de modèles d’assemblage, ainsi que les alignements des séquences aIF1 d’archées ont permis de proposer des régions ou acides aminés pouvant être impliqués dans la liaison au ribosome et/ou dans la sélection des ARNt initiateurs lors de la formation du complexe de démarrage. Afin de pouvoir étudier l’implication de ces régions ou acides aminés, j’ai mis au point une méthode d’étude de la liaison d’aIF1 à la petite sous-unité du ribosome par anisotropie de fluorescence. Cette étude met en évidence deux résidus basiques d’aIF1 impliqués dans la liaison au ribosome. D’autre part, les rôles d’aIF1 dans la sélection du codon de démarrage de la traduction et dans la stabilisation du complexe de démarrage sur l’ARNm sont étudiés par la méthode d’empreinte du ribosome ou toeprint. / A cell is subjected to different internal and external stimuli and must adapt quickly to fulfill its functions. This involves a fine regulation of gene expression. This occurs at the transcriptional level, but also at the translational level. Translation has three phases: initiation, elongation and termination. Selection of the start codon and therefore the choice of the mRNA reading frame is performed during translation initiation. From a kinetic point of view, translation initiation is the rate limiting step. Thus, it appears as a prime target for translational control.In archaea, initiation of translation involves a macromolecular complex containing the small subunit of the ribosome, mRNA, an initiator methionyl tRNA (Met-tRNAiMet) and three initiation factors (aIF1, aIF1A and aIF2). Interestingly, each of three initiation factors has an ortholog in eukaryotes.Archaeal and eukaryotic tRNAi have highly conserved bases A1-U72, at the extremity of the acceptor stem. This base pair was shown to be important for discrimination of initiator tRNAs from elongator tRNAs. In addition, other studies suggest the importance of the geometry of the A1-U72 pair for the initiator identity of those tRNAs. At the beginning of my thesis, no structural information was available to explain how the characteristics of the A1-U72 pair were involved in the selection of the initiator tRNA. At first, my thesis work involved the construction of an E. coli strain using as only source of tRNAi, a bacterial variant of tRNA initiator (tRNAfMet) having a base pair A1-U72 (tRNAfMetA1-U72). The use of this strain allowed us to obtain large quantities of purified tRNAfMetA1-U72. In addition, the crystal structure of this tRNAfMetA1-U72 has been determined at 2.8 Å of resolution. An unusual arrangement of bases A1 and U72 was observed.All archaeal translation initiation actors being available in the laboratory, a study of the archeal translation initiation complex by cryo-electron microscopy was achieved. The study identified two conformations of the tRNAi. In the first complex, both conformations (IC0-Premote (5.3 Å resolution) and IC1-Pin (7.5 Å resolution)) allowed us to propose a model for the accommodation of the initiator tRNA during start codon recognition.Finally, I was also interested in the role of the aIF1 factor. Availability of 3D structures, assembly models and alignments of the archeal aIF1 sequences allowed us to identify amino acids or regions that could be involved in ribosome binding and/or in the selection of initiator tRNA. In order to study the involvement of these regions, I have developed a method to study the binding of aIF1 to the small ribosomal subunit using fluorescence anisotropy. This study highlights two basic residues of aIF1 involved in binding to the ribosome. On the other hand, the roles of aIF1 in the selection of the start codon and in the stabilization of initiation complex on the mRNA were studied by the ribosome footprint method also called « toeprint ».