22455 research outputs found
Sort by
Accelerated Bayesian inference techniques for event based datasets
Get PDFThe updating of our knowledge in response to new information can prove to be complex, both in everyday life and in science. Bayes' theorem provides a systematic, probabilistic framework for updating our knowledge given new data. In any case of Bayesian inference - parameter estimation, model comparison, or field inference - probability distributions are used to encode our knowledge about the quantity of interest. This quantity is referred to as the signal. To compute the posterior probability of the signal in the presence of new data, Bayes' theorem combines prior knowledge of the signal with knowledge of the data measurement process, represented by the likelihood. This probabilistic approach not only provides a posterior estimate of the signal, but also allows for uncertainty quantification.
This thesis focuses on Bayesian field inference built upon information field theory. Inferring a field from an inherently finite data set is an under-constrained problem. Hence, the inclusion of prior knowledge is essential. In the present work, the implementation of generative, non-parametric prior models allows to exploit possibly complex and a priori unknown correlation structures of the signal during inference. Two distinct applications of Bayesian field inference are discussed: Bayesian evidence calculation and imaging.
The methodological part addresses the calculation of the posterior normalization - the Bayesian evidence. The evidence plays an important role in Bayesian model comparison. However, the evidence may be computationally intractable, for example due to complex relationships between the prior and posterior. Nested sampling provides a numerical estimate of the evidence by examining the likelihood as a function of enclosed prior volumes. In particular, the algorithm is based on statistical estimates of the prior volumes, introducing a stochastic error. For this reason a one-dimensional Bayesian field inference problem is formulated to obtain improved estimates of the prior volumes and the corresponding evidence.
The second and main part is dedicated to advancing Bayesian imaging of the X-ray sky. Observed by space-based telescopes, X-rays allow us to study some of the most energetic phenomena in the universe. However, the interpretability of the data is limited by overlapping X-ray sources, Poisson noise, and instrumental effects such as the point spread function. In this context, Bayesian forward models are constructed for the X-ray telescopes Chandra and eROSITA, tailored to their instrumental properties, to obtain denoised, deconvolved, decomposed, and spatio-spectral resolved images of the X-ray object of interest. The work presented includes the development of the JAX-accelerated open source software package J-UBIK to support future and existing Bayesian imaging models. Imaging results for the supernova remnant of SN1006 and for the Large Magellanic Cloud provide a detailed view of the diffuse X-ray structures and the separated point sources. This paves the way for future analysis of such fine-scale structures, such as shock fronts of supernova remnants, the construction of point source catalogs, and possibly improved instrument calibration
Vergleich zweier Line-Immunoassays zum Nachweis von spezifischen Antikörpern sowie Etablierung und Testung eines Line-Immunoassays zum Nachweis speziesspezifischer Antikörper gegen Erreger des Borrelia-burgdorferi-sensu-lato-Komplexes in caninen und equinen Seren
Get PDFDie Lyme-Borreliose ist weltweit eine der am weitesten verbreiteten durch Zecken übertragenen Krankheiten. Der heutige Goldstandard zum Nachweis der Borrelien-Infektion ist ein Zweistufenverfahren, in dem als erster Schritt ein kinetischer ELISA zur quantitativen Ak-Bestimmung und als zweiter Schritt ein LIA zur qualitativen Differenzierung positiver ELISA-Ergebnisse durchgeführt wird.
Ziel dieser Studie war es, zwei verschiedene LIAs hinsichtlich ihrer Handhabung und Resultate zu vergleichen, und somit eine Optimierung der Lyme-Borreliose-Diagnostik bei Hunden und Pferden zu ermöglichen. Hierfür wurden 200 canine und 200 equine Seren im betriebseigenen KELA, sowie mit LIA A und LIA B auf Borrelien-spezifische Ak untersucht.
Der LIA A erleichtert im Vergleich zum LIA B einige Arbeitsschritte und erlaubt eine automatische Auswertung der Ergebnisse mithilfe eines Scanners, welcher jedoch mit hämolytischem Serum Probleme hatte. Auch scheint der cut-off des LIA A für equine Seren etwas zu niedrig angesetzt zu sein, da fast die Hälfte der 200 Proben als seropositiv diagnostiziert wurden. Für den Einsatz bei geimpften Pferden gibt es bei beiden LIAs Verbesserungsbedarf. Im Fall von LIA A war vor allem die Interpretation der Reaktion der OspA-Bande als infektionsspezifisch kritisch anzusehen. Mittlerweile wurden beim LIA A Änderungen in Bezug auf hämolytisches Probenmaterial und die Rolle des OspA vorgenommen.
Der LIA B liefert weniger falsch-positive Ergebnisse und ist vor allem in Bezug auf equine Proben zuverlässiger in der Befundung und hier demnach die bessere Option. Bei der Untersuchung caniner Proben liefern beide Tests zufriedenstellende Ergebnisse. Aufgrund des Scanners und der automatischen Auswertung des LIA A im Vergleich zur visuellen Diagnosestellung des LIA B ist Ersterer bei einer Probenevaluierung mit wissenschaftlichem Hintergrund das Mittel der Wahl, da die Auswertungen leichter vergleichbar und statistisch erfassbar sind. Zudem kann menschlichen Fehlern vorgebeugt werden.
Der Spezies-LIA wäre eine nichtinvasive, schnelle und demnach für die Praxis wertvolle Methode der Differenzierung von Erregern des Bbsl-Komplexes. Die Vermutung, dass wie beim Menschen auch bei Tieren mehrere Spezies als nur Bbss zu klinischen Anzeichen einer LB führen, könnte dadurch bewiesen werden. Aufgrund der Kreuzreaktionen bei den caninen und murinen Kontrollseren, sowie der nicht möglichen Überprüfung der Bande von B. garinii, konnten die entstandenen Ergebnisse nicht verifiziert werden. Da diese Methodik der Diagnostik interessante neue Sichtweisen auf canine und equine LB liefern könnte, sollte der Spezies-LIA unbedingt weiterentwickelt und optimiert werden
Identification of pharmacological compounds selectively acting on the kinases-coupled channels TRPM6 and TRPM7
Get PDFGeneration and characterization of primate iPSCs for cross-species comparisons
Get PDFFor centuries, scientists have been intrigued by the phenotypic similarities and differences between humans and our closest living relatives, the primates. This has led to the development of various approaches aiming to understand the underlying genotype-phenotype relationship. Starting with investigating DNA sequence divergence between humans and chimpanzees, increased sequencing and advances in computational methods have enabled researchers to identify variants that are associated with human-specific phenotypes. In contrast, sequence conservation across multiple species with a close phylogenetic relationship can help to identify and interpret functional genetic variants. However, as changes in protein-coding sequences alone cannot account for the striking differences in phenotypes between species, gene expression regulation is considered as a pivotal contributor. Thus, the study of gene expression in humans and other primates can reveal unique insights into the evolution of gene regulation and can be used to identify genes and pathways that are associated with human-specific adaptations.
However, a major obstacle to comparative studies, especially those involving primates, is the availability of comparable samples. This is mainly due to ethical and practical reasons that complicate the acquisition of cells, particularly during developmental stages. To overcome these limitations, induced pluripotent stem cells (iPSCs) can be used, as they can be grown indefinitely in culture while maintaining their pluripotency. Thus, iPSCs can be differentiated into almost any cell type of the adult body, thereby allowing the study of rare and otherwise inaccessible cell types. Nevertheless, one of the major challenges in generating iPSCs from primates is obtaining the primary cells to be reprogrammed. To this end, I contributed to establishing a method that uses urine as a completely non-invasive cell source for deriving primate iPSCs. In the study, we demonstrate that urine-derived stem cells (UDSCs) can be isolated even from small volumes of primate urine and that the addition of a broad-spectrum antibacterial agent prevents contamination due to the unsterile collection from the zoo floor. Using this method, we were able to isolate and efficiently reprogram UDSCs from human, gorilla, and orangutan, thereby contributing to expanding the number of available cell lines from great apes for cross-species comparisons. To facilitate the use of this method by other researchers, I created a detailed protocol outlining the most crucial steps. These include the isolation and expansion of UDSCs from urine samples, as well as their reprogramming using a commercially available Sendai Virus (SeV) kit. In addition to the protocol, the manuscript also includes a video demonstrating how the most critical steps are performed in the lab. Moreover, we provide some best practices and troubleshooting guides that will enable a broader community to apply our method to their species of choice.
In recent years, not only great apes but also Old world monkeys have been extensively studied for the generation of iPSCs. Although iPSC lines exist for most of these primates, the number of different individuals and clones per species is still rather limited. However, given the high variability in gene expression among individuals and clones of the same individual, it is crucial for comparative primate genomics to increase the sample size to as many individuals or clones as possible. In this context, I generated new iPSC lines from one rhesus macaque, one baboon, and two vervet monkey individuals. To do this, I reprogrammed skin fibroblasts on feeder cells using a footprint-free SeV reprogramming method and later gradually transitioned the iPSCs to feeder-free culture conditions. We further characterized all iPSC lines using human iPSC characterization standards to prove their pluripotency and validate their undifferentiated state. As all of these cell lines can be cultured under feeder-free conditions in commercially available medium, this enhances their value for cross-species comparisons.
The invention of single-cell RNA-sequencing (scRNA-seq) technologies to measure gene expression, together with the possibility to generate iPSCs from primates that can be differentiated to nearly any desired cell type, have enabled the implementation of huge evolutionary studies. One example for such a study are scRNA-seq CRISPR inference (CRISPRi) screens to analyze molecular phenotypes across species. To this end, I contributed to generating human, gorilla and cynomolgus iPSCs that carry a doxycycline-inducible KRAB-dCas9 construct. The cell lines exhibited comparable down-regulation of target genes and comparable phenotypic effects in a scRNA-seq CRISPRi screen. Thereby, I helped to provide a valuable resource for performing CRISPRi in various primates, which can offer unique insights into human biology and evolution. In addition, comparative primate genomic studies can help to elucidated the evolutionary forces that drive the conservation or divergence of gene expression levels between species. In this regard, I used scRNA-seq to quantify gene expression during differentiation of our primate iPSCs towards embryoid bodies (EBs), a model for early development. I sampled single cells from ten different clones from four species at two time points, corresponding to differentiating and terminal cell types. This resulted in a comprehensive dataset comprising 85,000 cells, which we used to identify diverse cell types across all three germ layers. To accurately compare cell types across species, we developed a semi-automated computational pipeline combining classification and label transfer across clusters to identify orthologous cell types. This approach allowed us to investigate cross-species reproducibility of marker genes, revealing that human markers were less effective in macaques and vice versa. Furthermore, we found transcription factors to be the most conserved markers, highlighting their potential for cross-species studies. Overall, our study enhances the understanding of conserved and diverged molecular features in early primate development and we provide a well curated cell type reference for future in vitro studies.
In summary, within my thesis I contributed to the field of comparative primate genomics, by assisting in the development of a method to generate iPSCs from urine as a non-invasive cell source, and by providing a detailed protocol on how to apply this method. Moreover, I generated iPSC lines from fibroblasts of rhesus, vervet and baboon and contributed to the implementation of CRISPRi screens from primates. Additionally, I demonstrated, through a comparative differentiation approach to EBs, how orthologous cell types can be identified between species, thereby establishing a foundation for the identification of human-specific adaptations in gene regulation
Rational design of 1,2-dichalcogenides unlocks selective probes for the thioredoxin system
Get PDFVicinal dithiol oxidoreductases maintain redox homeodynamics by facilitating electron transfer, modulating protein structure and activity, and driving signaling processes. Despite their importance, the Trx/TrxR and Grx/GSH/GR systems lack high-quality turnover probes capable of dynamically monitoring their activity in live cells. This thesis addresses this gap by integrating fundamental knowledge of Trx/GSH system biology & biochemistry with state-of-the-art chemical probe research and basic principles of chalcogenol/dichalcogenide exchange (CDE) chemistry.
Dynamic imaging of enzyme activity requires probes that reversibly react with the target protein’s active site. To achieve this for the Trx/GSH systems, the turnover probes developed herein feature an artificial dichalcogenide substrate that is selectively reduced by the target reductase. Upon reduction, the probe undergoes irreversible fragmentation, leading to the accumulation of a fluorescent signal. The rates of the individual steps leading to signal generation govern a probe’s reactivity in biological systems, and achieving selectivity requires the design of artificial substrates whose CDE kinetics favour reduction exclusively by the reductase of interest.
This thesis pioneers the rational design of cyclic dichalcogenides for cellular dithiol reductase probes, resulting in practical syntheses of unprecedented cyclic 1,2-dichalcogenides, robust design principles for redox probe development, and the creation of cellular probes for Trx and TrxR.
Given the limited synthetic precedent for cyclic 1,2-dichalcogenides, a major focus was the development of efficient synthetic routes to these understudied motifs. Key achievements include regioselective syntheses of 1,2-thiaselenane amines and diastereoselective access to piperazine-fused 1,2-dithianes, 1,2-thiaselenanes and 1,2-diselenanes. Each sequence was designed for scalability & modularity, minimising the need for chromatographic purification.
Thioredoxin (Trx) is the most potent dithiol reductase in cells (katt, kfull↑). To achieve selective reduction by Trx, we designed cis-fused bicyclic disulfides that are thermodynamically and kinetically (kretro↑↑) equipped to resist reduction by cellularly abundant monothiols and weaker dithiol reductases (Grxs). Piperazine annellation ensured rapid fluorophore release rates (k’cyc), giving C-DiThias as the first cellular probes that selectively report on thioredoxin activity.
Thioredoxin Reductase (TrxR) is a unique selenolthiol reductase, which we targeted by using a 1,2-thiaselenane-4-amine substrate. Incorporation of selenium enabled high katt for TrxR’s selenolate while maintaining rapid reversion rates through intramolecular SN2 at Se (kretro). Kinetic selectivity for TrxR against thiol nucleophiles was reached since this kretro outcompeted the smaller katt for thiolates, and since full reduction from B is uniquely hindered for monothiols via iterative SN2 at the more electrophilic Se. As a result, probe RX1 fully resists >1000-fold challenge with GSH, reacts rapidly with 1/1000 equiv. of TrxR and has excellent TrxR-selective performance in cells, validated by knockout, selenium starvation, knockin, and chemical inhibitors.
Overall, this thesis presents a systematic approach to tuning individual probe activation rates through rational molecular design – an area that has received limited attention in the literature thus far. However, the multistep-reactive probe design strategy may prove valuable beyond Trx/GSH system probes: in developing chemical biology reagents or drugs that effectively distuinguish between on-target and off-targets, to achieve unmatched selectivity in complex cellular settings
In vivo analysis and therapeutic targeting of calcium-mediated axonal degeneration in a multiple sclerosis model
Get PDFMultiple sclerosis (MS) is a chronic inflammatory, demyelinating and degenerative disease of the central nervous system (CNS), characterised by the formation of focal inflammatory lesions and additional damage due to smouldering low-grade CNS inflammation. The clinical manifestation of relapsing-remitting MS – the rapid appearance of new neurological symptoms – is likely caused by acute inflammation and demyelination. The extent of long-term, irreversible disability, however, is determined by neurodegeneration, especially loss of axons. This degeneration is observable from the earliest stages of the disease and causes progressive disability once a critical compensable threshold is passed. Currently available immunomodulatory treatments for MS suppress relapses but cannot completely halt neuroaxonal degeneration. The development of primary neuroprotective treatments is therefore an important unmet clinical need.
In vivo imaging in experimental autoimmune encephalomyelitis, an animal model of MS, is a powerful tool to study the pathogenesis of axonal degeneration. Such experiments have shown that initial axonal damage is still reversible and that the likelihood of fragmentation of damaged axons is determined by intra-axonal calcium, which enters through “nanoruptures” in the plasma membrane. As some axons can re-establish calcium homeostasis spontaneously, we hypothesised that therapeutic interventions, which support axons in regaining calcium homeostasis will shift the balance in favour of recovery and reduce axonal loss. The aim of this thesis was therefore to explore calcium modulation for axoprotective therapy by (1) exploiting endogenous calcium homeostatic pathways and (2) developing an exogenous calcium chelation approach, selectively targeted to injured axons, to prevent side effects on physiological calcium signalling.
The basis of the selective targeting approach was to deliver a prodrug calcium chelator via the membrane “nanoruptures” that characterise damaged, high-calcium axons. As a proof-of-principle we first delivered fluorogenic compounds across damaged plasma membranes, thereby developing a membrane damage sensor, that is applicable across different cell types and in different types of membrane damage. The cytosolic localisation of the biosensor also makes it useful for targeting axons, which contrasts with nuclear DNA-binding stains such as propidium iodide. First steps of translating these findings into prodrug development have also been taken. To enhance endogenous calcium buffering, we performed experiments of rAAV-mediated, neuron-specific overexpression of the cytosolic calcium-binding proteins calbindin and calretinin. Despite optimisation attempts, the axonal expression achieved with this approach was low and did not reduce intra-axonal calcium levels or the degree of axonal loss in an MS animal model. Further improvement of the overexpression approach will be necessary to assess the axoprotective potential of calcium-binding protein overexpression.
In summary, this thesis lays foundational work to address the suitability of calcium as a neuroprotective target by developing a selective targeting approach to membrane-damaged axons and by evaluating an experimental approach for rAAV-mediated axonal overexpression of calcium-binding proteins.Die Multiple Sklerose (MS) ist eine chronische, entzündliche, demyelinisierende und degenerative Erkrankung des zentralen Nervensystems (ZNS), welche durch die Entstehung fokaler Entzündungsherde sowie durch eine progrediente diffuse Gewebsschädigung aufgrund eines niedrig-gradig schwelenden Entzündungsprozesses charakterisiert ist. Es wird angenommen, dass die klinische Manifestation der schubförmig remittierenden MS – die akut auftretenden neurologischen Ausfälle – durch die Entzündung und Demyelinisierung ausgelöst werden. Das Ausmaß der langfristigen und irreversiblen Behinderung hingegen wird durch die Neurodegeneration bestimmt, insbesondere durch den Verlust von Axonen. Diese Degeneration ist in den frühesten Stadien der Krankheit beobachtbar und verursacht eine fortschreitende Behinderung, sobald eine kritische, kompensierbare Schwelle überschritten wird. Die derzeit verfügbaren immunmodulatorischen Therapien für MS unterdrücken Krankheitsschübe, können aber die neuroaxonale Degeneration nicht vollständig aufhalten. Die Entwicklung primär neuroprotektiver Therapien ist daher dringend notwendig.
In vivo mikroskopische Untersuchungen der experimentellen autoimmunen Enzephalomyelitis, einem Tiermodell der MS, sind ein hilfreicher Ansatz, um die Pathogenese der axonalen Degeneration zu verstehen. Experimente dieser Art haben gezeigt, dass die anfängliche Schädigung der Axone noch reversibel ist und, dass die Wahrscheinlichkeit der Fragmentierung geschädigter Axone von der Konzentration des intra-axonalen Calciums abhängt, welches durch „Nanorupturen“ in der Plasmamembran eintritt. Da manche Axone ihre Calcium-Homöostase spontan wiederherstellen können, stellten wir die Hypothese auf, dass therapeutische Interventionen, welche Axone darin unterstützen ihre Calcium-Homöostase wiederherzustellen, die Balance zugunsten einer axonalen Erholung verschieben und den Verlust von Axonen verringern können. Das Ziel dieser Arbeit war es daher die Modulierung von Calcium als mögliche axoprotektive Therapiestrategie zu untersuchen. Dies erfolgte durch zwei verschieden Ansätze: (1) Durch die Ausnutzung endogener Mechanismen der Calcium-Homöostase und (2) durch die Entwicklung eines exogenen Calcium-Komplexierungsansatzes, welcher selektiv auf geschädigte Axone abzielt. Dies sollte Nebenwirkungen auf physiologische Calcium-Signale vermeiden.
Grundlage dieses Ansatzes war es, einen Calcium Chelator als Propharmakon durch jene „Nanorupturen“ einzuschleusen, welche geschädigte Axone mit erhöhtem Calcium charakterisieren. In einer Machbarkeitsstudie schleusten wir dazu zunächst fluorogene Stoffe über die geschädigte Plasmamembran ein und entwickelten so einen Membranschädigungs-Sensor, der in verschiedenen Zelltypen und für verschiedene Arten der Membranschädigung einsetzbar ist. Die zytosolische Lokalisierung des Biosensors macht ihn auch für die Darstellung von Axonen nützlich, im Gegensatz etwa zu Propidiumiodid, einem Farbstoff, welcher lediglich die DNA im Zellkern färbt. Erste Schritte zur Übertragung dieser Erkenntnisse auf die Entwicklung eines Calcium-puffernden Propharmakons wurden ebenfalls unternommen. Um die endogene Calcium-Pufferung zu unterstützen, wurden zudem Experimente zur rAAV-vermittelten, Neuron-spezifischen Überexpression der zytosolischen Calcium-bindenden Proteine Calbindin und Calretinin durchgeführt. Trotz mehrerer Optimierungsversuche war die so erreichte axonale Überexpression jedoch gering und konnte weder die intra-axonale Calcium-Konzentration noch den Verlust von Axonen in einem Tiermodell der MS reduzieren. Eine weitere Verbesserung des Überexpressions-Ansatzes wird notwendig sein, um das axoprotektive Potential Calcium-bindender Proteine zu untersuchen.
Zusammenfassend lässt sich sagen, dass diese Arbeit grundlegende Vorarbeiten zur Entwicklung neuroprotektiver Therapiestrategien, die an den erhöhten intraaxonalen Calciumkonzentrationen ansetzen, geleistet hat. Dies tut sie durch die Entwicklung einer selektive Zulieferungsmethode in membrangeschädigte Axone sowie durch die Evaluation eines experimentellen Ansatzes für die axonale rAAV-vermittelte Überexpression Calcium-bindender Proteine
Towards a mechanistic understanding of sensorimotor control and symptom perception in persistent physical symptoms
Get PDFDistressing physical symptoms that persist for months are frequent, occur across all areas of medicine and strongly impact quality of life. The association with measurable and reproducible pathophysiological processes is often loose or even absent and for most persistent physical symptoms (PPS), positive diagnostic markers are lacking, which challenges diagnosis and treatment. This thesis aims to contribute towards a better mechanistic understanding of PPS that can inform treatment and diagnosis by investigating symptom perception and sensorimotor processing in two examples of PPS, i.e., functional dizziness and post COVID-19 condition. We adopt a Bayesian brain perspective that proposes that the brain infers the most likely causes of sensory inputs by inverting an internal model that constitutes a probabilistic mapping between different states and sensory input as well as prior knowledge about these states. Recent theories have proposed that erroneous internal models can lead to the emergence of symptoms and dysfunctional motor processing, also in the absence of pathophysiological processes. Here, we provide further evidence in support of this hypothesis for functional dizziness and post COVID-19 condition. Using two different experimental paradigms, we were able to show that sensorimotor deficits (in functional dizziness) and increased breathlessness perception (in post COVID-19 condition) do not reflect altered and potentially pathological body states but rather are due to involvement of incorrect internal models. We highlight that different mechanisms could underlie these results and discuss the role of incorrect but highly precise priors in functional dizziness and maladaptive cost-functions in patients with post COVID-19 condition. In addition, we bridge the gap between experimental data and theories by developing a mathematical model that proposes a potential mechanism of how processing of respiratory data can lead to the emergence of breathlessness perception. In summary, this thesis provides an explanatory framework, a measurable marker of incorrect internal model use and an improved mechanistic understanding for functional dizziness and post COVID-19 condition. These findings can contribute towards development and refinement of existing treatments and reduce stigmatization of PPS