The transmission of different phenomena takes place when individuals interact in ways relevant to the specific transmission route. The definition of "contact" varies depending on the modelled phenomenon. For instance, sexual encounters are relevant for the transmission of sexually transmitted diseases (STDs), while close proximity is significant for airborne infections. To improve the accuracy of the simulation of the spread of phenomena of different origin, researchers have advanced various network models in previous studies. These models aim to better capture the dynamics of disease transmissions and their relationship to contact patterns.

In cases where the contact dynamics occur at a much slower pace than the spreading dynamics, leading to transmission whenever contact is made, it is sufficient to focus exclusively on the contact dynamics. If the network undergoes gradual changes and remains mostly unaffected throughout an outbreak, approximating it as a static network would be suitable. However, there are specific circumstances in which it becomes vital to consider both the dynamic nature of the contact network and the spread of the phenomena. Such consideration becomes particularly relevant when these two processes unfold at comparable timescales, and network modifications can shape the trajectory of the spread. In such circumstances, it is vital to include these changes in the analysis to obtain a comprehensive understanding of the dynamics of the spread. This research work addresses a specific scenario in which the temporal processes of the contact changes and the spreading process are closely interconnected. In this scenario, the occurrence of spreading events directly influences the structure of the network, subsequently influencing the subsequent spread of the disease. This adaptive behaviour enables a more realistic representation of behavioural changes that arise when individuals become aware of their infection and make choices such as self-isolation. This approach allows to capture the interplay between contact patterns and the progression of the spreading process, providing valuable insights into how different phenomena propagate within a population.

The necessity of this work arises from the critical need to balance accuracy and computational efficiency in simulating spreading processes on adaptive, time-evolving networks. The challenge in simulating transmission models on time-evolving adaptive networks stems from the stochastic nature of both spreading processes and contact behaviour. Choosing an appropriate stochastic simulation algorithm is challenging due to this dual stochasticity. While a range of stochastic simulation algorithms exists, selecting a suitable method is not straightforward. Approximate algorithms offer rapid computation but may compromise simulation accuracy and predictive reliability. Conversely, exact simulation algorithms yield accurate predictions but can suffer from computational inefficiency and protracted simulation times. This often hampers research progress and limits the generation of a sufficient amount of simulation trajectories for robust predictions. A distinct category of algorithms is available that allows for the explicit integration of internal dynamics and the evolution of contact network structures within simulations. However, these algorithms predominantly lack the capability to incorporate adaptive responses, a critical aspect of dynamic and responsive modelling in complex systems.

This thesis presents the development and validation of a novel hybrid algorithm, bridging gaps in current methodologies by combining the exact simulation of spreading dynamics with faster, either exact or approximate, simulations of contact dynamics. This methodology focuses on accurately simulating and predicting spreading dynamics while maintaining reliable statistics of contact behaviour, significantly enhancing computational performance for real-world scenarios.

Die in dieser Arbeit zusammengeführten Publikationen illustrieren in anschaulicher Art und Weise wie TZR-basierte Immuntherapien einen immer größeren Stellenwert in der Therapie von malignen Erkrankungen einnehmen. In den ersten Arbeiten wurde sehr deutlich der Stellenwert der Immunpeptidomik unterstrichen, welche vor allem durch technische Weiterentwicklungen in der Isolation und Analyse von HLA Liganden die eigene Sensitivität weiter verbesserte. Hierdurch wird die Immunpeptidomik zum idealen Werkzeug, welches tumorspezifische Targets definieren kann, indem es nicht nur die Abundanz von HLA Liganden abschätzt, sondern auch deren Veränderungen nach unterschiedlichen biologischen oder pharmakologischen Stimuli nachverfolgen kann. Dieser Aspekte wurde im dritten und vierten Manuskript sehr klar erläutert, da es hier gelang das Immunpeptidom mittels epigenetischer Modulatoren so zu verändern, dass es die Immunevasion umkehren konnte und neue Subgruppen von tumorspezifischen HLA Liganden definierte. Für Tyrosinkinaseinhibitoren dagegen konnten wir zeigen, dass deren spezifische Inhibition eines Signalweges auch zu einer definierten Präsentation von Peptiden aus dem Pathway nach sich zieht. Darüber hinaus liefert die Immunpeptidomik auch wertvolle Informationen über potenzielle off-Targets mit deren Hilfe die Spezifität einer TZR-basierten Therapie genauer eingeschätzt wird, um höhergradige Toxizitäten zu vermeiden. Daher ist der Einsatz der Immunpeptidomik bereits heute eine unumgängliche Methode, welche die Entwicklung eines jeden TZR-basierten spezifischen Therapeutikums begleiten sollte wie die erfolgreiche Umsetzung u.a. in meinen Arbeiten zur NDC80 spezifischen TZR-imitierenden CAR-T Zelle bereits belegt hat. Hier wurde erstmals systematisch ein tumoragnostischer HLA Ligand unabhängig von der Art des Quellproteins identifiziert, welcher eine spezifische Elimination von Tumorzellen erlaubt, obwohl NDC80 auch in gesunden Zellen nachgewiesen werden kann. Das große therapeutische Fenster durch die unterschiedliche Prozessierung in Tumorzellen und dem Normalgewebe erlaubt somit die Nutzung eines bisher unbeachteten Targets, welches erst durch die Immunpeptidomik klar definiert werden konnte. 88 Der Stellenwert der Immunpeptidomik wird vermutlich in den kommenden Jahren noch weiter zunehmen, da durch die immer bessere Sensitivität der Massenspektrometrie und optimierte Isolationsmethoden die benötigten Zellzahlen für derartige Analysen immer kleiner werden und somit auch immunpeptidomische Analysen aus zirkulierenden Tumorzellen schon bald möglich sein könnten. Kurzum, die massenspektrometrische Analyse von HLA Liganden ist und bleibt der Goldstandard für die Entwicklung TZR-basierter Immuntherapien und sollte bei der Entwicklung dieser Therapeutika immer integriert werden. Die Erfolge der präklinischen, translationalen und klinischen Forschung der vergangenen Jahre zu welchen auch meine Arbeiten entscheidend beigetragen haben, belegen, dass die Zulassung weiterer TZR- basierter Wirkstoffe inklusive der TZR-imitierenden CAR T Zellen für die Behandlung maligner Erkrankungen nur eine Frage der Zeit ist.

Volcanic eruptions pose a grave risk to metropolitan areas and the global economy. Improving the accuracy, reliability, and warning time of eruption forecasts opens avenues to make the lives of some 600 million people living at active volcanoes safer. An effective way of achieving these ends is diversifying current monitoring strategies. This thesis employs changes in the seismic wave propagation velocity (dv/v), retrieved from the ubiquitous ambient seismic noise, as a new measure to quantify alterations in the dynamics of the volcanic system. We explore an easily automatable method to quantify seismic velocity changes from ambient seismic noise, which we implemented into the software suite SeisMIC. SeisMIC offers end-to-end processing of ambient seismic noise in an easy-to-use, well-documented, modular, and adaptable Python software suite. Compared to other ambient seismic noise software solutions, SeisMIC offers improved computational performance and is particularly suited to satisfy the extraordinarily variable processing needs for ambient noise volcano monitoring. We apply this method to two datasets at active volcanoes: The Klyuchevskoy Volcanic Group (KVG), in Kamchatka, Russia, and Mount St. Helens (MSH), in the United States of America. At the KVG, we analyse data recorded during the KISS project, which consists of one year of seismic data recorded by over 110 stations. On Kamchatka, the constant volcanic tremor causes fluctuations in the wavefield, violating the assumption of source stationarity in our method. We address these fluctuations using a hierarchical clustering algorithm to identify times with stationary noise fields, for which we can create shorter dv/v time series. Coincident with the first inflation at Bezymianny volcano and 11 months before its eruption, we observe a dv/v increase. Furthermore, we find evidence for a higher damage susceptibility of volcanic rock following the M7.2 Zhupanov earthquake and, with the help of a simple empirical model, link changes in rainfall and snow depth to dv/v changes. The long-standing seismic network at MSH has recorded continuous seismic waveform data for over 25 years. We use this unique dataset to investigate changes in the seismic velocity during, before, and after MSH’s 2004-2008 eruption period. Over time, new seismic stations are deployed, while old ones are removed, destroyed, or upgraded. This poses a particular challenge when creating long-term dv/v time series as the normalisations for time series of station combinations recording during different times diverge. We overcome this challenge using an innovative approach that predicts dv/v at new station combinations from data from pre-existing stations and their spatial sensitivity kernels. The average dv/v time series exhibits two marked increases. Together with the first explosions of the 2004-2008 eruption and a GNSS downward movement, we find a sudden velocity increase caused by the pressure drop and unplugging of the volcanic conduit. Between 2017 and 2020, we detect a largescale velocity increase below the volcanic edifice focused on the location of MSH’s plumbing system. From the dv/v increase, a temporal gravity anomaly, and the GNSS observations, we infer a slowing or stopping of MSH’s post-eruptive reinflation during this interval. This thesis paves the way towards automated volcanic eruption forecasting using seismic velocity changes by distributing a software suite well suited to monitor seismic velocity changes in a breadth of environments in real-time and by demonstrating that the seismic velocity can be employed to quantify volcanic dynamics at various active volcanoes, even when technical and environmental conditions are unideal. Future work should focus on implementing seismic velocity change monitoring at volcano observatories and strengthening the physical links between various environmental, tectonic, and volcanic forcings to the seismic velocity.

Instances of geometric problems usually have both a concrete geometric and a more abstract representation. Given the geometric representation it is often easy to find the abstract one while the converse is not true: Given an abstract representation it can be challenging to find a geometric one or decide that no such representation exists. How challenging it is generally depends on the complexity of the given abstract representation. In this thesis we study two different problems with the aforementioned properties:

Weak Unit Disk Contact Representations. A unit disk contact representation (UDCR) of a graph is a set of interior-disjoint unit disks in the plane (each disk corresponds to one node) such that two disks touch if and only if their corresponding nodes have an edge. The notion of weak unit disk contact representations (weak UDCRs) weakens this condition and only enforces that two disks must touch if the corresponding nodes have an edge. If two nodes don’t have an edge, their corresponding disks are still allowed to touch. The problem comes in two flavors: the first are graphs without embedding where the neighboring disks can be arranged in any order. Here we show that the problem is NP-hard for trees and provide a linear time algorithm for caterpillars, which are trees that become paths when all leaves are removed once. The second flavor are graphs with a fixed embedding where a given order of the neighboring disks in a weak UDCR must be respected. Here we show that the problem is already NP-hard for general caterpillars. We also show that we can decide in linear time whether a caterpillar has a weak UDCR that can be placed on a triangular grid and whose disks for the path of inner nodes are strictly x-monotone.

Colored Nearest Neighbor Graphs. In the second part of this thesis we look at a one-dimensional geometric problem. Here we are given a set of one-dimensional points and a list of line segments between neighboring points such that every point has at least one incident line segment. We then assign a non-empty set of colors to each point and for each assigned color create an edge in this color between the point and the closest point that also has this color. A valid assignment of colors to the points has two properties: First, every created edge corresponds to a line segment between the same endpoints and for every line segment there exists such an edge. Second, there can be no two edges with different colors between two points. It is easy to see that such a valid color assignment always exists: select a unique color for each line segment and assign this color to both endpoints. In this thesis we answer the question of how many colors are needed for a given input and how long it takes to find a valid assignment. We show that for both one and two colors we can decide whether a valid assignment exists (and also find it) in linear time. To this end we make some structural observations that help us narrow down the possible color assignments that we need to look at. There are two defining structures responsible for increasing the necessary number of colors: local maxima which are line segments that are longer than both adjacent line segments, and small gaps which are missing line segments between two points such that the missing line segment is not longer than both adjacent line segments. We show that in the worst case the number of local maxima increases the number of colors logarithmically and the number of small gaps increases them linearly. Finally, we provide a dynamic program that, given an input and a number of colors, finds a valid color assignment with this number of colors or returns that no valid color assignment exists. The running time of this algorithm is exponential in the number of colors.

This thesis investigates the reasons behind divergent policy responses of Peru and Colombia to Venezuelan forced displacement. Specifically, it aims to explain why two countries with similar histories of international migration, political and economic systems, and relations with the sending country, have adopted different approaches to a comparable phenomenon in terms of demographic composition, scale and duration. Under President Iván Duque, Colombia introduced the Estatuto de Protección Temporal para Migrantes Venezolanos (ETPV) in 2021, an unprecedented liberal policy which provides Venezuelans with a 10-year stay permit and a pathway to permanent residency. In contrast, Peru, under President Martín Vizcarra, shifted from an initially welcoming stance to a more restrictive approach by imposing a visa requirement known as the Humanitarian Visa on Venezuelans in 2019 to control their entries.

To address the reasons behind these divergent policy responses I employ outcome-explaining process-tracing and comparative analysis in a Most Similar Systems Design. I triangulate data from 65 interviews with policy-makers and experts, surveys, legal documents, public statements from parliamentary debates and media, and secondary literature. I analyse this data using abductive Thematic Analysis based on a codebook.

I argue that the contrasting policies in these countries primarily stem from variations in their policy-making processes. The study identifies relevant differences in the authorities involved – whether politicians or bureaucrats –, the nature of political considerations – domestic versus foreign affairs – and the presence or absence of cost-benefit analyses. My findings show that Colombia’s adoption of the ETPV resulted from a policy-making process that involved both politicians and bureaucrats, prioritised foreign affairs considerations and included a comprehensive cost-benefit analysis. In contrast, Peru’s Humanitarian Visa was formulated by a more limited group of political actors, including the President and key ministers, with minimal bureaucratic input, focusing on domestic concerns and lacking an informed cost-benefit analysis.

The study also highlights the importance of political and institutional contexts in shaping variations in policy-making. In Colombia, foreign affairs considerations are prevalent due to the salience of Venezuela in its political agenda – rooted in historical ties and the role of Venezuela in Colombia’s armed conflict –, as well as the executive’s damaged international reputation resulting from its stances on the armed conflict. Conversely, Peru’s focus on domestic concerns is driven by the executive’s vulnerability amid a political crisis and its strained relations with the parliament. Furthermore, Colombia’s well-established institutional capacities, developed through prolonged conflict and internal forced displacement crises, enabled a robust bureaucratic response and an informed cost-benefit analysis. In contrast, Peru, lacking similar institutional experience and facing political instability, exhibited weaker bureaucratic capacities, leading to a more politically driven approach.

This research advances the forced displacement and migration policy fields by examining the role of policy-making, including the involved authorities and their considerations, and of bureaucracies and knowledge in forced displacement policies in the Global South – aspects thus far largely overlooked in the existing literature. While supporting established theoretical assumptions regarding the impact of the relations between sending and receiving countries and domestic concerns, the study provides new insights into why certain considerations – whether related to foreign or domestic affairs – are prioritised in specific contexts, beyond factors tied to the nature of the exodus itself. Furthermore, it explores how determinants deemed important by the literature, such as regime type, executive vulnerability, and previous institutional frameworks, interact to shape policy. The study introduces the concept of established institutional capacities as a crucial determinant of policy-making, expanding the understanding of how political instability and historical legacies affect policies on forced displacement. Departing from studies in other regional contexts, it reveals the limited role of social contexts and ethnic kinship in explaining policy differences within Latin America. By drawing parallels with countries of the Global North and contrasting them with those of the Global South, my results challenge the traditional South/North divide prevalent in the literature.

Finally, my research suggests several implications for future policy-making on forced displacement. Centralising migration responsibilities within a single department, incorporating trained bureaucrats, and relying on evidence-based practices can enhance policy effectiveness. Furthermore, increased international collaboration, learning from international experiences and establishing stable, legally enshrined policies can address the complexities of forced displacement more effectively. This perspective not only advocates for policies that respect migrants’ rights but also enhances positive outcomes for receiving countries and societies.

Adoptive chimeric antigen receptor (CAR)-T cell therapies have led to tremendous clinical success, especially in treatment of hematological malignancies. However, optimizations are still required to tackle remaining challenges such as antigen escape, on-target off-tumor toxicity, limited availability of tumor-specific antigens as well as insufficient T cell persistence and tumor infiltration. A significant level of expectation is currently placed in transgenic T cell receptors (TCRs), which feature a highly sensitive, naturally evolved signaling machinery and cover a broader range of targets. This work aimed at combining both technologies, CAR and transgenic TCR, to increase the anti-tumor response, to facilitate dual-targeting, to minimize the risk for antigen escape and to potentially achieve synergistic effects. As proof of concept, initial experiments were focused on the combination of the clinically well-established second-generation CD20-CAR and the previously published transgenic dNPM1-TCR. CAR’TCR-T cells, co-expressing CD20-CAR and dNPM1-TCR, revealed increased anti-tumor activity upon dual stimulation, not only upon short-term co-culture but also upon long-term repetitive in vitro stimulation. Accordingly, CAR’TCR-T cells displayed a unique transcriptomic signature, indicating increased T cell activation and proximal signaling. In-depth characterization of CAR’TCR-T cell functionality also included the analysis of alternative costimulatory domains in the CAR designs, displaying no significant difference between 4-1BB- and CD28-costimulated CAR’TCR-T cells. Moreover, it was addressed whether co-expression of CD20-CAR promotes functional recovery of the MHC class I-restricted dNPM1-TCR in CD4+ T cells. Finally, dual-specificity and enhanced anti-tumor activity of CAR’TCR-T cells were verified in vitro and in vivo for a clinically relevant AML setting by co-expressing dNPM1-TCR and CD33-CAR. Especially the treatment with Triple-T cells, meaning a cellular composition of CAR’TCR-T, CAR-T and TCR-T, led to promising results in vivo, demonstrating increased anti-tumor cytotoxicity compared to a mixture of CAR-T and TCR-T cells. In summary, this work supports the approach of co-expressing a CAR and a transgenic TCR to achieve dual-specificity and enhanced T cell anti-tumor activity. The superior performance of Triple-T highlights the therapeutic benefit of CAR’TCR-T cells and the potential clinical applicability through co-transduction with two different lentiviral vectors.

Die Diagnose neurologischer Erkrankungen bei Vögeln ist oft herausfordernd. Die neurologische Untersuchung stellt einen wesentlichen diagnostischen Schritt dar, der zur Lokalisation der symptomauslösenden Läsion innerhalb des Nervensystems führt und damit die Basis für die Einschätzung möglicher Differenzialdiagnosen bildet. Allerdings sind Daten bezüglich der Anwendbarkeit neurologischer Tests bei Vogelarten nur sehr begrenzt in der Fachliteratur verfügbar. Darüber hinaus sind klinische Beschreibungen von Krankheitsbildern bei Wildvögeln oft nur spärlich vorhanden. Das Ziel dieser Dissertation bestand darin, die neurologische Untersuchung sowie die Diagnosestellung von Erkrankungen mit neurologischen Symptomen bei Vögeln zu verbessern. In der ersten Studie wurden physiologische Reflexantworten und Reaktionen im Rahmen der neurologischen Untersuchung klinisch gesunder Individuen verschiedener Vogelarten ermittelt. Bei 42 Tauben (Columba livia domestica), 42 Höckerschwänen (Cygnus olor) und 42 Greifvögeln, darunter zwölf Mäusebussarde (Buteo buteo), 24 Turmfalken (Falco tinnunculus) und sechs Habichte (Accipiter gentilis), wurde eine neurologische Untersuchung durchgeführt. Eine Reihe von neurologischen Tests konnte artenunterschiedlich als geeignet für die untersuchten Spezies identifiziert werden. Bei Tauben und Höckerschwänen wurden bei fast allen Haltungs- und Stellreaktionen reproduzierbar zuverlässige Reaktionen beobachtet. Im Gegensatz dazu waren die Haltungs- und Stellreaktionen der Hintergliedmaßen bei Greifvögeln aufgrund von Abwehrverhalten oft nicht durchführbar. Die Überprüfung der Kopfnerven und der meisten spinalen Reflexe ergaben artspezifisch variable Ergebnisse. Insbesondere der Gastrocnemiusreflex war bei keinem Individuum der Studie auslösbar. Die Ergebnisse deuten darauf, dass es vogelartspezifisch sowohl Unterschiede in der Durchführbarkeit der neurologischen Untersuchung als auch in der Auslösbarkeit und Ausprägung der Antworten im Rahmen der verschiedenen Tests gibt. In einer weiteren Studie wurden die klinischen, pathologischen und virologischen Aspekte von Infektionen mit dem West-Nil-Virus (WNV) bei zehn freilebenden Habichten aus Deutschland beschrieben. Das WNV ist ein zoonotisches Arbovirus, das erstmals 2018 in Deutschland nachgewiesen wurde und zu einer neuroinvasiven Erkrankung mit erhöhter Mortalität bei Vögeln, Menschen und anderen Säugetieren führen kann. Habichte sind besonders empfänglich für eine klinische Erkrankung und gelten als Indikatorart für das Auftreten des WNV in Europa. Aus diesem Grund sind Daten über das klinische Erscheinungsbild von großer Bedeutung für die frühzeitige Identifizierung von Verdachtsfällen. Zwischen Juli und September 2019 wurden insgesamt zehn freilebende Habichte (8/10 waren juvenil und 9/10 weiblich) in die Klein- und Heimtierklinik der Freien Universität Berlin verbracht und umfassend untersucht. Die klinischen Befunde, die im Laufe des stationären Aufenthaltes erhoben wurden, waren Apathie und Inappetenz (10/10), ein schlechter Ernährungszustand (10/10), eingeschränkter Visus ohne okuläre Befunde (4/7), Uveitis (3/7), Hyphäma (1/7) und eine Hornhauterosion (1/7). Alle Tiere entwickelten im Verlauf neurologische Symptome, darunter Stupor (3/10), Krampfanfälle (3/10), Kopftremor (2/10), Kopfschiefhaltung (2/10), Monoplegie eines Beines (2/10) und Ataxie (2/10) und eine neurologische Untersuchung ergab eine multifokale neuroanatomische Lokalisation (8/9) oder eine fokale Lokalisation auf das Großhirn (1/9). Zusätzlich wurden Aerosacculitiden oder Pneumonien (7/10), traumatische Verletzungen (3/10) wie z.B. Frakturen der Carina sterni, klinisch apparente Infektionen mit Eucoleus spp. und Trichomonas spp. (3/10) sowie Myiasis (2/10) festgestellt. Trotz symptomatischer Behandlung verschlechterte sich der Zustand der Vögel, was zum Tod oder zur Euthanasie bei einer medianen Überlebenszeit von zwei Tagen führte. Die häufigsten histopathologischen Befunde waren eine Meningoenzephalitis (9/10), Myokarditis (8/10), Iridozyklitis (8/8) und Myositis (7/10). Die WNV-Infektion wurde mittels quantitativer Echtzeit-Polymerase-Kettenreaktion mit reverser Transkription (RT-qPCR) in Organproben (10/10) und Blut (6/8) diagnostiziert und durch serologische und immunhistochemische Untersuchungen bestätigt. Die Ergebnisse dieser kumulativen Dissertation beschreiben artspezifische, physiologische Reflexantworten und Reaktionen im Rahmen der neurologischen Untersuchung bei verschiedenen Vogelarten. Diese sollten bei der Interpretation neurologischer Tests bei den entsprechenden Spezies beachtet werden, um die Untersuchung möglichst aussagekräftig und stressarm für den Patienten zu gestalten. Dieser Untersuchungsgang konnte in einem weiteren Teil der kumulativen Dissertation bei Habichten mit neurologischen Symptomen ebenfalls angewendet und eine, in Deutschland relativ neue, Infektionserkrankung durch das WNV bei zehn freilebenden Habichten umfassend klinisch, pathologisch und virologisch charakterisiert werden. Diese Erkrankung sollte bei Habichten mit entsprechenden neurologischen Symptomen, insbesondere im Sommer, als wichtige Differenzialdiagnose Berücksichtigung finden. Insgesamt tragen diese Untersuchungen zur besseren Diagnosestellung neurologischer Erkrankungen bei Vögeln bei.

Die vorliegende Arbeit untersucht den wiederkehrenden Figurentyp der christlich konnotierten Erlösungsfigur sowie die Darstellung entsprechender Erlösungsvorstellungen im Science Fiction-Film seit 1990. Dabei stellt sich die Frage, inwiefern die Erzählmuster der Science Fiction mit den in der westlichen Gesellschaft dominanten christlichen Vorstellungen von Erlösung ineinandergreifen, sich bedingen und schließlich als archetypische Figuren und narrative Elemente verstanden werden können. Auch die audiovisuelle Darstellung wird daher untersucht, um wiederkehrende Muster aufzudecken. Aus diesem Grund greift die Arbeit vorrangig auf Methoden der Figuren- und Szenenanalyse zurück, betrachtete jedoch auch den diegetischen Rahmen und seine Parallelen zu biblischen Texten. Untersucht werden dazu vorrangig westlich produzierte Filme. Zudem stellt sich die Frage, welchen Zweck diese Einbindung religiös geprägter Vorstellungen und Motive in den Unterhaltungsfilm erfüllt. Die Arbeit zeigt somit, dass christlich konnotierte Erlösungsfiguren sowie die Darstellung von Erlösungsvorstellungen im Science Fiction-Film als nahezu archetypisch zu verstehen sind, auch wenn selten von Seiten des Films ein direkter Bezug zur christlichen Vorstellung hergestellt wird. Der Film nimmt durch die Nutzung dieser religiös geprägten Motive und Figuren insofern eine wesentliche Rolle für das Publikum ein, dass er mit dem hoffnungsvollen Ende, das er darstellt, die Diskrepanz zwischen einem mangelhaften Ist- und einem erwünschten Soll-Zustand überbrückt. Was für viele Jahrhunderte etablierte Kirchen und Gebete für die Menschheit geleistet haben, sucht sie heute zunehmend in individuellen spirituellen Vorstellungen und, wie sich durch die Einbindung der Erlösungsvorstellungen in den Science Fiction-Film zeigt, auch in den (Unterhaltungs-) Medien.

Quantum error correction is an essential ingredient in the development of quan- tum technologies. Its core subject is to investigate ways to embed quantum Hilbert spaces into a physical system such that this subspace is robust against small imperfections in the physical systems. This task is exceedingly complex: for one, this is due to the vast diversity of possible physical systems with dif- ferent inherent structure to use. For another, every different physical setting also comes with different types of dominant imperfections that need to be protected against. Bred by the complexity of this technological ambition, research on quantum error correction has developed into a large field of research that ranges from questions about the engineering of small systems with a single photon to the creation of macroscopic topological phases of matter and models of complex emergent physics. A quintessential tool in quantum error correction is the stabilizer formalism, which tames complicated quantum systems by enforcing symmetries. A Gottesman-Kitaev-Preskill (GKP) code is a stabilizer code that creates a log- ical subspace within the infinite dimensional Hilbert space of a collection of quantum harmonic oscillators by endowing it with translational symmetries. While practical approaches to GKP codes consider the infinitude of the Hilbert space, as well as the infinitude of the translational symmetry group as obstacles for implementation, in theory these are precisely the features that make the theory of GKP codes particularly rich, well behaved and well-connected to fascinating topics in mathematics. The purpose of this thesis is to explore these connections: to understand the coding theoretic- and practical properties of GKP codes, utilizing its rich mathematical foundation, and to provide a foundation for future research. Along this journey we discover – through the looking glass of GKP codes – how quantum error correction in general fits into a fabulous mathematical world and formulate a series of dreams about possible directions of research.

Multiple sclerosis (MS) is a chronic inflammatory disease of the central nervous system (CNS) that is characterized by autoimmune-driven inflammation, resulting in myelin destruction and damage to neuronal axons and cell bodies. These alterations manifest as CNS lesions in magnetic resonance imaging (MRI). In MS patients, MRI has shown the presence of cerebral gadolinium (Gd) retention following repeated injections of mainly linear gadolinium-based contrast agents (GBCAs). Using the experimental autoimmune encephalomyelitis (EAE), a mouse model of MS, we previously observed that EAE mice exhibited enhanced cerebellar short-term Gd retention compared to healthy controls (HC) following injections of gadopentetate dimeglumine. Since MS patients generally undergo multiple contrast-enhanced (CE) MRIs throughout their lives, the concern of augmented Gd retention and potential long-term consequences may be specifically relevant for them. This thesis aimed to investigate the long-term character of inflammation-promoted brain Gd retention and contributing factors after repeated application of linear gadopentetate or macrocyclic gadobutrol. By combining longitudinal T1 mapping of the brain and cross-sectional (laser ablation-) inductively coupled plasma mass spectrometry (LA-/ICP-MS) assessments of blood and cerebellum, we took an interdisciplinary approach to address Gd retention patterns and kinetics in EAE and HC mice. MRI calibration was performed to assess the impact of inflammation on Gd bonds and compare the binding behavior of the two tested GBCAs. In addition, this thesis presents the establishment of an ex vivo model of living brain tissue and its utilization to determine Gd-induced neurocytotoxicity under naïve and inflammatory conditions. Within this model, the retained Gd content was visualized by Imaging Mass Cytometry (IMC), and quantified by ICP-MS. Quantitative T1 relaxometry and LA-ICP-MS results confirmed that in EAE mice Gd retention was enhanced after injections of both GBCAs. While gadopentetate led to persistent retention, particularly inside the inflamed cerebellar nuclei (CN), gadobutrol caused weak and diffuse Gd content, which cleared over the observational period. MRI calibration showed that in vivo MRI underestimated the T1 effect of Gd after the administration of gadopentetate, suggesting that Gd may be partially bound in insoluble deposits. After incubation with gadopentetate ex vivo, retained Gd content at approximately 1.5 mM led to a reduction in neuronal viability, which further decreased under inflammatory conditions. Taken together, the results indicate that inflammation promotes the extent of Gd brain retention, particularly after linear GBCA administration, and significantly influences its retention pattern, binding to endogenous ligands, and cytotoxic potential. Thus, multiple administrations of mainly linear GBCAs may pose a long-term risk for patients suffering from chronic neuroinflammatory disorders like MS.

This thesis focuses on the development of FTJ devices using a Hf0.5Zr0.5O2 (HZO) ferroelectric layer, demonstrating their compatibility with CMOS integration and their applicability in neuromorphic hardware. A bilayer structure comprising metal- ferroelectric-dielectric-metal layers with a ∼ 10 nm HZO ferroelectric layer and a thin tunneling Al2O3 layer has been investigated.

We examine the impact of dielectric positioning, metal electrode placement (with W and TiN), and dielectric thickness on device performance. Additionally, we explore the role of charge traps in the dielectric or at the dielectric-ferroelectric interface, and the influence of the fabrication process on charge trap density and polarization switching behavior. W bottom electrode is found to give optimized device performance, and positioning Al2O3 next to the bottom electrode further enhances the device performance in terms of ON current and ON/OFF ratio. Furthermore, longer pulses are necessary to stabilize higher remnant polarization due to charge trap dynamics.

We also analyze various electrical parameters affecting FTJ device performance, demonstrating that the cycling waveform significantly influences the wake-up process and the resulting remnant polarization in TiN-Al2O3-HZO-W FTJ devices. Square waveforms outperform triangular waveforms, yielding higher remnant polarization (PR) post-wake-up. By employing an asymmetric waveform for field cycling and adjusting the pulse width, the PR and the ON/OFF ratio after wake-up are significantly improved.

Finally, we explore the integration of bilayer FTJ devices into CMOS back-end-of- line (BEOL) processes, demonstrating a 1T1C circuit by connecting an FTJ in the BEOL with an nMOS transistor in the front-end-of-line. Measurements on standalone FTJ devices in the BEOL reveal an ON/OFF ratio of 18 and an ON current density of 24.5 μA/cm2. Crucially, BEOL fabrication has negligible impact on transistor characteristics, and the 1T1C circuit exhibits a 2.6-fold amplification of the FTJ ON current. The FTJ devices integrated on the CMOS-BEOL demonstrate multiple resistance states with the application of partial switching Reset and Set pulses. These FTJ devices have the potential to be utilized in neuromorphic hardware systems.

The fascinating vision of molecular nanotechnology lies in the use of single molecules as machines or devices and the controlled assembly of molecular matter at the nanometer scale (nanomanufacturing). The possibility to manipulate and control things on a small scale was first introduced by Richard Feynman in his famous 1959 speech “There’s plenty of room at the bottom”. Resulting applications, like electronic circuits, sensors or nano-machines, would, due to their small dimensions, open the door to completely new technologies, with advantages in cost, operation speed (efficiency) and power dissipation. In contrast to top-down approaches (e.g. by improving lithographic techniques), the use of molecules reveals several important advantages: 1. Molecules have nano-scale dimensions by nature: Thus, the use of single molecules automatically leads to structural dimensions at the atomic scale. The top-down miniaturization of existing applications and devices requires fundamental improvements of conventional fabrication techniques, which are limited. 2. Molecular recognition: Intermolecular interactions can be used to grow molecular nanostructures. Molecules are known to self-organize according to their chemical properties and to self-assemble in topologies that reflect the interactions between them. By controlling the chemistry of the involved molecules, various structures of different size and shapes can be produced. 3. Capabilities of organic chemistry: By choosing the structure and composition of a molecule, chemical synthesis can be used to produce molecules of precisely defined properties. 4. Functions: Molecules can have specific functions, for instance switching between different states (of characteristic optical, magnetic and electronic properties). In fact, the latter is a fundamental concept in nature, where many molecules have multiple stable isomers.

PSA (tapes) are permanently sticky adhesives that can be applied by slight pres-sure to a large variety of surfaces. This easy application has led to the wide-spread use of PSA not only in the DIY sector but also in the construction, rail-way vehicle, and automotive industries, where the fire behavior of the materials is of crucial importance. The fire behavior of PSA-bonded materials is therefore a research topic of high interest to industry and academia. Due to the viscoelas-tic state of PSA, it is a challenge to develop well compatible flame retardants that do not bloom out or migrate to the adhesive surface and degrade the adhe-sion between the substrate and the adhesive. To overcome this problem, phos-phorus flame retardants, as a very versatile class of flame retardants that can be modified to meet the specific demands of the polymer matrix, are a good choice for a PSA matrix. A further advantage of phosphorus-based flame retardants is that they are effective at low concentrations, allowing the PSA tapes to keep their mechanical properties. They act in several modes of protection, depending on the chemical environment of phosphorus within the flame retardant and the polymer matrix in which they are used. The state of art is to develop phosphorus-based flame retardant PSA tapes that perform well in flammability tests and expect them to have beneficial influence in the bonded product, without this relationship being proven. The questions of how different flame retardants work in PSA and PSA tapes and how these flame retardant adhesives influence the burning behavior of bonded materials is elab-orated on in this doctoral dissertation. In the first step of this dissertation, the fire behavior and pyrolysis of commer-cially available phosphorus-based flame retardant adhesives (tapes) were ana-lyzed and compared to adhesive tapes without flame retardant to gain a basic understanding of how adhesive tapes behave in fire and how flame retardants affect this behavior. Fire behavior analysis was performed in different fire sce-narios and bond designs established in a multi-methodical approach. The reac-tion of the adhesive tape to a small flame was investigated as a free standing object, bonded one-sided onto different substrates, and as sandwich like bonds (substrate/tape/substrate). The interaction between flame retardant, adhesive matrix, carrier, and substrate was explained by systematically examining the pyrolysis of the adhesive, the flammability of the tapes, and the fire behavior of bonded materials in an ignition and a developing fire scenario. Py-GC/MS, TGA FTIR, PCFC and hot-stage FTIR measurements were used to identify the mode of action of the flame retardant and its interaction with the PSA matrix. The flammability tests, UL 94, OI, and the single-flame source test identified the effect of the flame retardant on the flammability of the tape and the effect on the flammability of bonded materials. Measuring and comparing the fire be-havior of the bonded and monolithic materials in the cone calorimeter led to an understanding of how PSA tapes influence the fire behavior of bonded sub-strates. The fire behavior of different monolithic and bonded substrates with a wide spectrum of burning characteristics was investigated to gain information on the substrate specificity of the tape influence. The second step of the dissertation consisted in preparing differently phospho-rus-based flame retardant PSA and PSA tapes. One predominantly gas phase ac-tive, one condensed phase active and one covalently bonded phosphorus flame retardant were used in a PSA matrix which mainly consisted of poly(n-butyl acrylate). Determining the adhesion and cohesion properties of the PSA tapes guaranteed the possible application as PSA tape. Their flame retardant mecha-nism and decomposition and combustion behavior were analyzed thoroughly in Py-GC/MS, TGA FTIR, hot-stage FTIR and PCFC to gain precise insights into the chemical and physical mechanisms that govern the pyrolysis process. The dif-ferent modes of actions and mechanisms were subsequently connected to the flammability of adhesive tapes which were prepared from the synthesized flame retardant PSA and assessed in UL 94 and OI. In the last step of this systematic doctoral dissertation, the self-prepared adhe-sive tapes were used to bond different substrates whose fire behaviors were then investigated in different fire scenarios simulating ignition, developing and fully developed fire. This made it possible to connect the pyrolysis mechanism, mode of action, and fire behavior of the free standing adhesive tapes with the fire behavior of bonded materials. Different carriers (AL and PET) were used in different substrates (wood, bisphenol-A polycarbonate, polymethyl methacry-late) to yield in an understanding of the influence of the adhesive tape carrier on the burning behavior of the individual substrates. The first step shows that phosphorus flame retardants significantly improve the flammability of the commercially available adhesive tapes. The gas phase active flame retardant resulted in a UL 94 V-2 rating and a large increase (5.3 vol.-%) in OI, and significantly changed the burning behavior in the single-flame source test. Tape-bonded materials behaved substantially different from their mono-lithic counterparts in cone calorimeter measurements, where the adhesive tape mostly acted as an insulating barrier separating the bonded substrate layers. This separation led to new fire risks in PC and increased the FIGRA up to 20% and the PHRR up to 26%. Cone calorimeter measurements also showed that phosphorus flame retardants and carriers must be tailored to each other because they can react and degrade the protective properties of the carrier, as it was the case for the AL carrier in combination with a commercial phosphorus flame re-tardant. The strong enhancing effect of the flame retardant on the flammability of the non-bonded adhesive tapes was not present in adhesive tape-bonded ma-terials where the fire behavior was determined by the substrates. During the second step of this dissertation, the individual decomposition of the PSA matrix and the flame retardants were analyzed in Py-GC/MS, TGA FTIR and PCFC and detailed decomposition and flame retardant mechanisms were postu-lated. The gas phase active DOB 11 released PO and PO2 radicals at low temper-atures resulting in a V-2 UL 94 rating at small concentrations. The condensed phase active flame retardant, RDP, improved the charring of the PSA tape sur-face, as it is suspected to be a precursor of phosphoric/ polyphosphoric acid. These acids lead to elimination reactions which result in unsaturated structures and finally enhanced char formation. The small amount of RDP had, normalized on the p-concentration and compared to the other flame retardants, the biggest positive effect on the OI of the adhesive tape. DOPO-pentyl-methacrylate alone has a low decomposition temperature (starting at 300 °C) but was covalently bonded to the polymer backbone and therefore decomposed together with the polymer (starting at 350° C). It released PO and PO2 radicals that improved the flammability in all fire tests. In addition to its beneficial influence on the fire behavior, the covalently bonded flame retardant also increased the mechanical properties at elevated temperatures, which is an important parameter for PSA tapes. The pyrolysis and flammability of these different flame retardant types for PSA show that the phosphorus flame retardants have individual advantages and need to be tailored to the matrix and the application. The third and final step of this doctoral dissertation focusses on the fire behav-ior of self-prepared adhesive tapes with different flame retardant PSA and dif-ferent carriers used to bond substrates representative of automative, railway vehicle, and construction applications. Cone calorimeter measurements showed that it is not the adhesives and the flame retardants in the PSA matrix that result in changes in the fire behavior, but rather the choice of carrier. In the case of PMMA, for example, an AL carrier can improve the fire behavior by acting as a barrier to protect the underlying material, leading to a 25% reduction in MARHE and a 30% reduction in PHRR compared to a PET carrier. The same AL carrier used in PC substrates resulted in a 30% increase of MARHE and FIGRA, indi-cating higher fire hazards and poorer performance in the cone calorimeter test. To determine the transferability of the results obtained for adhesive tapes to other adhesives, the fire behavior of adhesive tape-bonded materials was com-pared to materials bonded by liquid thin-layer adhesives. Similar effects of the liquid adhesives were obtained in the cone calorimeter, suggesting that the find-ings for the burning behavior of adhesive tape bonds are likely to be general-izable to materials that are bonded by other adhesives. Overall, a fundamental understanding of the fire behavior of adhesive tapes and bonded materials was generated and the individual effect of different flame re-tardants for PSA tapes was investigated thoroughly. Based on the new findings, the state of the art PSA development is questioned and new, more end applica-tion focused research is suggested. The strong impact of adhesive gaps in several tape-bonded materials provides a promising outlook for future research in-to the fire behavior of materials bonded by other adhesives and using different substrates and applications.

Photon upconversion is a non-linear process where multiple photons are absorbed, and a single photon with higher energy than the incident ones is emitted. This process is often achieved by lanthanide ions, such as Erbium ions which are generally embedded in optical inert crystals like Sodium-Yttrium-(tetra) Fluoride. The ability of the Erbium ions to covert the near-infrared photons into shorter infrared, visible and ultraviolet wavelengths opens up many applications such as life sciences and photovoltaics. The efficiency of the photon upconversion through Erbium ions strongly depends on Erbium ion’s ability to absorb near-infrared photons. Unfortunately, Erbium, like any other lanthanide group element, is not an efficient near-infrared absorber. Their small absorption cross-section requires intense laser excitation power densities to obtain considerable upconversion luminescence. This requirement limits its applicability for practical applications. In this thesis, an approach to enhance photon upconversion of the Erbium-doped and Erbium - Ytterbium co-doped Sodium-Yttrium-(tetra)Fluoride upconversion nanoparticles is studied by harvesting strongly enhanced near-fields on silicon metasurfaces. The aim is to reduce the required external excitation power densites for efficient photon upconversion luminescence and to gain a better understanding of resonant phenomena on silicon metasurfaces. To fulfil this aim, the resonant phonemonon on Silicon metasurface resulting strong near fields and photo-physical dynamics of the upconversion nanoparticles are meticulously investigated by experimental and theoretical methods. As an upconversion enhancing platform, large area (25 cm2) silicon mono- and double-layer metasurfaces with hexagonal lattice are produced by a nanoimprinting lithography based method. The resonant phenomena on silicon metasurfaces, such as leaky modes and bound states in continuum on silicon metasurfaces are probed by optical spectroscopy and aided with finite difference element simulations. In the light of the numerical calculations, the spectral position of the leaky modes of silicon metasurface is engineered by a single and simple production parameter;silicon metasurface thickness. The excitation power density-dependent upconversion dynamics of the upconversion nanoparticles in solution are examined. The results on excitation power density dependent measurements shed on light electonic transition dynamics and luminescence quenching processes. By placing the upconversion nanoparticles on the silicon metasurface and variation of the incident angle, the effect of the 1550 nm light coupling with leaky metasurface modes on upconversion luminescence is in- vestigated. The efficient coupling results in more than 2000-fold enhanced photon upconversion emission on silicon metasurfaces when it is compared with the emission of upconversion nanoparticles on a planar silicon surface. Moreover, upconversion of Erbium-Ytterbium co-doped upconversion nanoparticles in polymer matrix is studied. The upconversion of the polymer-upconversion nanoparticle layer are investigated with 980 nm laser excitation. The enhanced photon upcoversion on metasurface more than 1000-fold is observed under 980 nm excitation. These results pave the way to achieve efficient upconversion luminescence even at low excitation power densities inspiring applications in the fields of biophotonics and photovoltaics.

Cells as the fundamental units of life, exhibit significant complexity, hosting intricate networks of biochemical processes and structural components which enable them to perform a vast array of functions necessary for the survival and adaptation of living organisms. The dynamic and diverse nature of cells causes them to constantly interact and adapt to environmental cues. Cell types and morphologies vary widely among organisms and tissues, reflecting the discrete functions and specialized roles in maintaining the viability of the biosystems involved. Abnormalities in cell size and shape can be associated with pathological conditions, which can be monitored via imaging techniques and used for diagnostic purposes. During the processes of endocytosis and exocytosis, the plasma membrane can form invaginations and exhibit shape remodeling events. The ability of the plasma membrane to undergo dynamic morphological alterations allows cells to control cellular communication, recognition processes, adhesion and immune responses as well as the transport of substances between the cell interior and exterior environment. The plasma membrane composed of a diverse arrangement of lipids, proteins and carbohydrates, plays an important role in maintaining the structural integrity and metabolic activities of cells thus taking part in the regulation of the cellular homeostasis. However, the diverse components of the plasma membrane and their various roles in the extracellular and intracellular activities make them very challenging and complex to characterize.

The structural and functional complexity of the plasma membrane has required the construction of simpler membrane models for the investigation of membrane dynamics. These models can also characterize the individual effects and properties of membrane components during cellular interactions and transport processes. Model membranes can be manipulated with external stimuli. Among them, light irradiation offers several advantages as a non-invasive, reversible, biocompatible and facile tool to provide high spatiotemporal control in biomimicry of cellular events. One of the efficient approaches to utilize light for the manipulation of membrane models is to prepare biomimetic platforms with photoswitchable lipids which can reversibly change their molecular conformation upon light irradiation. In this doctoral thesis, we have developed a light-triggered, multifunctional, and smart biomimetic platform by using phosphatidylcholine (referred as POPC here and in the main text), one of the most abundant phospholipids in animal cells, combined with a photoswitchable azobenzene lipid analog (referred as azo-PC). This platform was designed to provide an optical control of the membrane properties, shape, and molecular transport in the biomimetic system through the photoisomerization of azo-PC under UV and blue light.

First, we comprehensively investigated the reversibility, kinetics and effects of photoswitching on the material properties of varying compositions of azo-PC containing minimalistic membrane models including giant unilamellar vesicles (GUVs) as minimal cell-size models, Langmuir monolayers, large unilamellar vesicles and supported lipid bilayers, and combined the results from a variety of experimental approaches to those obtained from molecular dynamics simulations. These investigations showed that azo-PC photoisomerization induces dynamic alterations in membrane properties, affecting bilayer packing, elasticity, and interleaflet interactions. Using a method based on vesicle electrodeformation and optical microscopy, we revealed how the photolipid, introduced at various fractions, alters the membrane area upon isomerization and found excellent agreement with simulations. UV illumination of azo-PC GUVs triggered trans-to-cis photoisomerization, resulting in a significant increase in membrane area and a ten-fold decrease in bending rigidity. Trans azo-PC bilayers were found to be thicker than POPC bilayers but exhibited higher specific membrane capacitance and dielectric constant. This suggests an enhanced ability to store electrical charges across the membrane. Furthermore, incubation of preformed POPC GUVs with azo-PC rendered them photoresponsive, suggesting therapeutic potential for optical control of cellular activities. By using a wide range of experimental and computational approaches, we collected accurate results about the characterization of the light-induced modifications in azo-PC containing membranes, which also allowed us to discuss the discrepancies in the previously reported values in the literature and explain the origins of these discrepancies.

Next, we demonstrated the application of photoswitching of azo-PC doped GUVs for the transport of protein-rich droplets by performing light-triggered endocytosis of biomolecular condensates. Protein-rich condensates are phase separated membraneless organelles acting as vessels for biochemical reactions in cells during important cellular processes including signal transduction and gene expression. In our studies, condensates were prepared from the plant protein glycinin, which is a prominent storage protein in soybeans. UV-light-induced trans-to-cis photoisomerization of azo-PC results in an instantaneous increase in vesicle area, which promotes the wetting of GUVs by condensates and their rapid endocytosis for an appropriate condensate-vesicle size ratio. The process is fully reversible by exposure to blue light, allowing precise spatiotemporal control of the condensate-membrane interaction. The affinity of the protein condensates to the membrane, and the kinetics, reversibility and degree (whether partial or complete) of the engulfment processes were quantified from confocal microscopy images. Theoretical estimations confirmed that the adhesion of protein condensates to azo-PC vesicles in cis conformation under UV irradiation is energetically favorable. Experimental results, in good agreement with theoretical estimations, demonstrated that light and azo-PC photoisomerization can be employed as a versatile system to modify membrane-condensate interactions in a fast and reversible manner. To the best of our knowledge, this is the first study in the literature to utilize photoisomerization to control the delivery process of a biomacromolecule across a minimalistic artificial cell, providing a promising approach for further exploration in the control of cellular transport of biomacromolecules.

Lastly, employing azo-PC photoisomerization we established optical control of the activity of mechanosensitive ion channels reconstituted in GUV membrane, which enabled the transport of small molecules across the membrane. These pore-forming transmembrane proteins can open and close in response to changes in membrane properties and tension. As a model protein, we used the bacterial mechanosensitive ion channel of large conductance (MscL) and reconstituted it into azo-PC containing GUVs. Labeling MscL with a dye allowed us to monitor the reconstitution process through confocal microscopy, determine critical parameters of the reconstitution process and develop a protocol for the reconstitution of MscL into azo-PC containing GUVs. In order to understand the initial conformation of the incorporated MscL in the GUV membrane, the vesicles were subjected to a permeability test by adding a water-soluble, membrane-impermeable sulforhodamine dye to the external medium of the GUVs. Most of the GUVs remained impermeable, indicating that the majority pf MscL in the membrane stayed in the closed state after reconstitution. This trend changed when UV/blue illumination was applied to the MscL-reconstituted azo-PC vesicles. UV/blue illumination altered the membrane properties of azo-PC doped GUVs and triggered the opening of the MscL channel, as monitored by the permeation of the sulforhodamine molecules across the GUV membrane. Our preliminary results showed that light can be used as an efficient tool to catalytically activate the MscL channel. Further studies should focus extensively on the optimization and characterization of MscL reconstitution and gating processes.

Overall, our findings in this doctoral thesis provide an essential background for understanding and optimizing light-triggered drug delivery platforms and photoregulation of shape-dependent cellular processes such as endocytosis, exocytosis and intercellular trafficking through azo-PC photoisomerization. Considering our results on the control of membrane shape, mechanics and trafficking by light irradiation, photoisomerization may be used as a promising biomedical alternative for cell repair processes. Photoswitchable biomimetic platforms can be further developed as light-triggered smart activators for high-precision regulation of cellular mechanisms.

Mosquito-borne alphaviruses like Chikungunya virus (CHIKV) and Mayaro virus (MAYV), which caused endemic outbreaks involving millions of patients over the past decades, pose an emerging threat due to climate-driven expansion of the virus vectors. Recently, the first vaccine against CHIKV was approved by the FDA, however, the lack of antiviral treatments underscores the need for a deeper understanding of virus-host interactions, potentially leading to the development of novel antiviral strategies. In this thesis, we investigate the role of ubiquitin-specific protease 10 (USP10) in the context of CHIKV and MAYV. Our findings demonstrate that USP10 overexpression significantly reduces infection rates for both alphaviruses. This antiviral effect is independent of USP10's deubiquitinase activity and instead relies on its interaction with the stress granule protein G3BP, mediated by a specific motif (FGDF) within USP10. Notably, USP10 overexpression disrupts the formation of stress granules during CHIKV infection, likely by sequestering G3BP and preventing its aggregation. Exploring the impact of USP10 on different stages of the viral replication cycle, our findings suggest that USP10 primarily targets the later stages, inhibiting the formation and release of new virus particles. In contrast, CHIKV glycoprotein-mediated entry and viral RNA replication are not significantly affected by USP10. Analogously to USP10, CHIKV nonstructural protein 3 (nsP3) binds G3BP via two FGDF motifs. Cells infected with G3BP binding-deficient mutants displayed no reduction in viral infection upon USP10 overexpression, underlining a critical role of the G3BP-nsP3 interaction in the antiviral effect of USP10. Finally, we explored the emergence of potential escape variants after serial passaging of CHIKV under selection pressure by USP10 overexpression. Notably, mutations were observed in the nsP2 and E2 proteins, suggesting potential rescue mutations in these proteins counteracting the antiviral activity of USP10. This highlights the selective pressure exerted by USP10, highlighting its importance as an antiviral factor. In conclusion, this thesis establishes USP10 as a potent cellular inhibitor of alphavirus infection. By targeting the G3BP-nsP3 interaction, USP10 offers a promising path for the development of novel antiviral strategies.

Spermatozoa are considered phenotypically equivalent because, during spermatogenesis, their precursor cells are organized in a syncytium where they share mRNA through intercellular bridges. This assumed phenotypic equivalence is challenged by a selfish genetic element encoded by the t-haplotype. The mouse t-haplotype is a naturally occurring variant of mouse chromosome 17 that shows high transmission from heterozygous (t/+) males due to a poison/antidote mechanism. t-Distorter genes exert detrimental effects on all sperm of these males. The t-responder, which can counterbalance these effects, remains restricted to the t-sperm, leading to their selective rescue and transmission ratio distortion (TRD) in favor of the t-chromosome. I first investigated whether energy metabolism contributes to the altered flagellar function, causative for t-haplotype TRD via the glycolysis enzyme phosphoglycerate kinase 2 (Pgk2). Pgk2 is overexpressed in mice with t-chromosomes. I generated male mutants carrying a t-specific Pgk2 knockout allele (tw5Pgk2∆/+) and showed that sperm of mutant tw5Pgk2∆/+ mice have increased progressivity, which suggests an improvement of wild-type sperm performance (which in t/+ are compromised by t-distorters). An important goal of this work was to evaluate whether TRD is a more widespread phenomenon and whether other genes, independent of the t-haplotype, can also cause TRD. Four candidate genes, Spata45, Tex29, Tex46, and Txndc2, were selected based on two criteria: late expression during spermatogenesis and minimal sharing among haploid spermatids. I generated reporter mouse lines of each gene and analyzed the distribution of their transcripts in the seminiferous tubules. Tex29 was shown to be restricted to the haploid spermatid encoding it. I then generated a Tex29 heterozygous loss-of-function mutant and showed that the transmission of the mutant allele was not affected. In addition, to study the function of Spata45, Tex29 and Tex46. I generated homozygous knockout mutants of these genes and examined whether they exhibit sterility or subfertility phenotypes. Spata45-/- and Tex29-/- males showed no fertility defects. However, Tex46-/- males are sterile demonstrating that Tex46 is essential for male fertility. Tex46-/- mice have normal testes, typical seminiferous tubule structures and can produce spermatozoa. However, the morphology of the sperm head reveals malformations affecting the arrangement of the plasma membrane, the internal and external acrosomal membranes, and the apical hook. Although Tex46-/- sperm cannot fertilize oocytes with zona pellucida, they successfully fertilize zona pellucida-free oocytes. It is proposed here that Tex46-/- sperm have an impaired acrosome reaction and are therefore unable to cross the zona pellucida. In this work a novel gene has been identified, that is critical for male fertility in mice, and may also play an important role in human male fertility.

In this work, a general and direct method for the synthesis of silver(I) perfluoroalcoholates has been investigated, including an examination of their structural properties in both solid state and solution, as well as an evaluation of their effectiveness as transfer reagents. The synthesis was performed by reaction of AgF with corresponding perfluorinated carbonyl compounds in acetonitrile (MeCN). The obtained silver(I) perfluoroalcoholates in MeCN were found to be stable at –18 °C over months. X-ray crystallographic analysis of the perfluoroalcoholate single crystals revealed a structure characterized by silver centers bridged by alcoholate ligands, while two alcohol ligands coordinate to one silver center. In acetonitrile solutions, Ag[OCF3] adopts a variety of structural forms as shown by IR spectroscopy. Furthermore, these silver(I) perfluoroalcoholates were found to be valuable as user-friendly transfer reagents, facilitating the synthesis of Cu[OCF3], Cu[OC2F5], [PPh4][Au(CF3)3(OCF3)], and various fluorinated alkyl ethers. Trifluoromethyl fluorosulfonate (CF3OSO2F) and trifluoromethoxysulfur pentafluoride (CF3OSF5), both bearing the ‒OCF3 moiety, were investigated regarding their dielectric behav-ior. Both compounds exhibit higher breakdown voltages compared to sulfur hexafluoride (SF6), with averaged relative breakdown voltages of 1.3 ± 0.2 for CF3OSO2F and 1.4 ± 0.2 for CF3OSF5, compared to 1.0 for SF6. This makes them promising candidates for dielectric appli-cations where higher voltage tolerance is necessary. During an electrical breakdown, both compounds, CF3OSO2F and CF3OSF5, decompose in a rate similar to that of the dielectric (CF3)2CFCN. The decomposition behavior was analyzed using IR spectroscopy and GC-IR techniques to identify the decomposition products. Moreover, the molecular structures of these compounds were successfully obtained using in situ crystallization methods. In addition, eval-uations of their physical properties, including vapor pressure, critical parameters, and melting points, were performed. These assessments further demonstrated the viability and perfor-mance of these materials as dielectric materials.

This work includes new methods for the introduction of fluorinated groups into organic molecules as well as investigations into photoredox catalysed C-F activation. The evaluation of benzothiazolium (BT) salts as deoxygenative trifluoromethylthiolation and fluorination reagents provided new insights into the Lewis acidity and mechanistic properties of BT-reagents. Both the polymerisation of tetrahydrofuran (THF) to trifluoromethylthioethers and the α-deprotonation of THF could be observed and investigated. BT-SCF3 could also be used for the synthesis of amides via acyl fluoride intermediates, which offers new application possibilities of the BT-reagents for fluorination reactions. The investigation of sulphur hexafluoride and bis(trifluoromethyl)peroxide (BTMP) as radical sources for pentafluorosulphanylation and trifluoromethoxylation reactions is also presented. BTMP can be used for the synthesis of α-OCF3 ketones and silyl enol ethers, as well as for the synthesis of allyl-OCF3 products starting from silylated compounds without any activators. In addition, studies on the Wagenknecht reaction with new reaction partners for the direct synthesis of SF5 compounds were presented and mechanistic properties of the reaction were identified based on the results. The thesis also discusses the photoredox catalysed reaction of α-fluoroacetophenones with silylated coupling partners, with the presented method showing new routes to polyfunctionalised synthetic building blocks.

The growing global demand for metals needed for the green energy transition has led to renewed exploration efforts. Ancient sedimentary basins contain a number of important resources such as zinc (Zn) and lead (Pb), and sometimes germanium (Ge), gallium (Ga), and indium (In). Some of the largest Zn deposits are called clastic-dominant (CD-type) deposits because they are hosted in clastic rocks like mudstones and siltstones. These deposits are thought to have formed when metal-rich fluids from deep in the Earth’s crust were expelled towards the surface along faults. However, many aspects of this overall model are not well understood. For example, the composition of the ancient metal-rich fluids is largely unknown, along with the mechanisms by which metals are deposited to form economic enrichments in certain locations.

The Selwyn Basin (Canada) contains a number of important CD-type deposits that formed hundreds of millions of years ago. The rocks hosting these deposits are now incorporated into the Canadian Cordillera, meaning they have been uplifted and deformed. This can make it challenging to interpret the rocks and understand how the deposit formation (mineralization) occurred. As a result, it is crucial to carefully examine samples that preserve some of the key primary features of the host rocks and the deposits.

This project evaluates aspects of the CD-type deposit model using various approaches that utilize two sample sets. 1) Barite- and pyrite-rich samples from the Late Devonian Canol Formation in Canada, which contain no Zn, Pb sulfides and formed at the same time as those rocks containing the deposits at other locations, were used to determine how the ancient environment was before the deposits formed. 2) Mineralized rock samples from a newly discovered CD-type deposit (Boundary Zone, Canada) were utilized to evaluate i) how these deposits formed, ii) the physicochemical properties of the metal-rich fluids, and iii) what essential metals are present. To answer some of the abovementioned questions, these two sample groups were used to make petrographic, mineralogical, and geochemical observations across various scales, from hand specimen to microscopic levels.

Data obtained through detailed petrographic and isotopic analyses indicate that the barite and pyrite in the Canol Formation formed during early diagenesis and that biological activity was critical for converting sulfate to sulfide. Similar mineral phases are observed in the samples from the Boundary Zone, where sulfide formed during early diagenesis likely reacted with metal-bearing hydrothermal fluids during an initial stage of ore formation. The first stage is dominated by fine-grained sphalerite formed as layers due to the replacement of quartz (and barite) components of the rocks. A second ore stage consists of several sphalerite types forming in cracks within the same rocks after fracturing. Critical metals, including Ge and deleterious components like Hg, occur in high amounts in the sphalerite from both the mineralized stages. Furthermore, experiments conducted on tiny droplets of fluids trapped within sphalerite and quartz indicate that the mineralizing fluids consist of variable salinity and homogenization temperature ranging from low (around 120 °C) during the early ore-forming stages to high (around 260 °C) at a later period. This suggests that the Boundary Zone deposit formed due to mixing of these fluids at some point.

Altogether, this thesis provides significant insights into components of the CD-type deposit model. It shows how vital microbial activities were during the formation of rocks that later host these deposits and also highlights prolonged hydrothermal fluid flow that could form multiple mineralization types. These findings are valuable for exploration strategies in the Macmillan Pass district and similar geological settings.