16580 research outputs found
Sort by
Structural analysis of Asgard archaeal chromatin
No full textThe origin of eukaryotic chromatin represents a pivotal question in understanding the evolution of cellular complexity. Chromatin organization underpins genome accessibility, transcriptional regulation, and cellular differentiation - features that distinguish eukaryotes from their prokary-
otic ancestors. Recent phylogenomic analyses identify the Asgard superphylum of archaea, including Lokiarchaeota, Heimdallarchaeota, and Hodarchaeota, among others, as the closest relatives of eukaryotes. These lineages therefore provide a unique opportunity to explore how primitive DNA-binding proteins may have evolved into the dynamic chromatin systems characteristic of eukaryotic nuclei.
This doctoral research presents the first high-resolution structural and mechanistic analysis of Asgard archaeal chromatin, focusing on the core histone protein HHoB from a representative Hodarchaeal lineage. Using cryogenic electron microscopy (cryo-EM) complemented by advanced biochemical and biophysical methods, the study reveals that Asgard chromatin does not exist as a single static entity, but rather as a dynamic ensemble capable of adopting two fundamentally distinct conformations. The first, a novel open chromatin assembly, exhibits a loosely packed, filamentous architecture in which the DNA remains partially exposed, allowing potential access by transcriptional or repair machinery. The second, a closed chromatin assembly,
forms a compact, highly ordered fiber stabilized by extensive inter-histone interactions. The open conformation displays structural parallels to the eukaryotic (H3–H4) octasome, and is possibly an Asgard-specific innovation toward greater chromatin flexibility. In contrast, the closed state recapitulates structural features seen in other archaeal chromatin systems, indicating evolutionary continuity across archaeal lineages. Biochemical and structural assays further reveal that divalent cations such as Mg2+ modulate the equilibrium between these states, with elevated Mg2+ concentrations favoring chromatin compaction - highlighting a possible ionic regulatory mechanism. This study provides the first structure-based model of Asgard chromatin organization, expanding our understanding of chromatin architecture in an evolutionary context
Solving Dilemma Games with Evolving Conditional Commitments
No full textI study a formal mechanism that can sustain Pareto optimality in a new and very broad class of dilemma games. In the absence of a central authority that could enforce multilateral agreements, the mechanism is based on binding unilateral commitments, which condition a player's (possibly multidimensional) contribution on other players' contributions. I show that unexploitable better response dynamics converge to Pareto optimal contributions when the game is played recurrently
Wrapping your brain (microvasculature) around cerebral malaria: A 3D brain microvessel model to explore pericyte alteration and angiopoietin/Tie-2 dysregulation
No full textCerebral malaria (CM) is a severe complication of Plasmodium falciparum infection associated with blood-brain barrier (BBB) breakdown. A hallmark of cerebral malaria is cytoadhesion of P.falciparum-infected red blood cells (iRBCs) to the brain microvasculature, yet the mechanisms behind CM pathogenesis and BBB breakdown remain largely unknown. Dysregulation of the endothelial-protective angiopoietin/Tie-2 axis is a predictive marker of CM-induced fatality and therefore is an attractive causative factor in the promotion of brain endothelial dysfunction observed in CM patients. In the brain microvasculature, the angiopoietin/Tie-2 axis consists of the endothelial receptor Tie-2 and its ligands, angiopoietin-1 (Ang-1) and angiopoietin-2 (Ang-2), secreted primarily by brain pericytes and brain endothelial cells respectively. However, previous human in vitro brain endothelial models to study CM have omitted brain pericytes and therefore, do not fully recapitulate this pathway.
Here, I developed a minimal 3D microfluidics-based brain microvasculature model that is composed of both primary human brain endothelial cells and brain pericytes, and replicates angiopoietin/Tie-2 axis signaling. I demonstrated that the model reproduces in vivo brain pericyte coverage of the microvessels, cell type-specific marker expression and ultrastructural brain endothelium-pericyte interactions. Perfusion with iRBC egress products increased microvessel permeability, albeit with minor changes in microvessel morphology and brain endothelium-pericyte interaction markers. Strikingly, intact iRBCs and iRBC egress products halted secretion of Ang-1 with the microvessels, revealing for the first time a role of brain pericytes in CM-associated angiopoietin/Tie-2 axis dysregulation. Pre-incubation with recombinant Ang-1 offered partial protection against microvessel barrier breakdown induced by iRBC egress products, highlighting reduction in Ang-1 secretion as a promoter of BBB breakdown during CM. On the contrary, Ang-2 secretion was not increased in the presence of intact iRBCs and their egress products suggesting involvement of additional CM-associated factors in angiopoietin/Tie-2 axis dysregulation. Use of a clinical therapeutic that enhances Tie-2 activity displayed rapid partial protection against the barrier disruptive effects of iRBC egress products, proposing interventions that restore homeostatic function of the angiopoietin/Tie-2 axis as potential adjunctive therapies against CM. Overall, this model provides evidence towards brain pericytes representing a key cell type in angiopoietin/Tie-2 axis dysregulation and BBB breakdown, while offering insight into novel therapeutic avenues in which to counteract CM pathogenesis
Atom Trap Trace Analysis of 39Ar: Method Development and Application to Glacier Ice Dating
No full textGlacier ice stores a unique range of past environmental signals and therefore serves as an invaluable archive of Earth’s climatic history. However, to fully exploit this
archive, the knowledge of the age distribution of the ice is key. Radiometric dating methods are an important addition to traditional age dating techniques like annual layer counting, especially when these fail due to discontinuities in the record.
This thesis aims to establish the 39Ar dating tracer as a well-understood tool for dating glacier ice. The advent of the atom-optical Atom Trap Trace Analysis (ATTA)
technology for 39Ar measurements has been associated with a significant reduction of required sample sizes and the tracer can therefore now routinely be applied to
glacier ice.
As a major part of this thesis, the ATTA laboratory in Heidelberg was improved to deliver higher atom count rates for high-quality measurements with large long-term
sample throughput.
Furthermore, two studies on Alpine glaciers are presented in this thesis, the summit glacier of Weissseespitze, where part of a 6000 a continuous record was dated, and the Jamtalferner, which is a much younger, highly variable low-altitude glacier. In both campaigns, 39Ar ages are complemented by other tracer ages from radiocarbon or tritium, emphasizing the importance of this tracer in covering the full age range found in Alpine glaciers.
The specific requirements for the application of 39Ar dating to glacier ice are investigated by comparing different ice sampling and sample preparation techniques. Additionally, the diffusion of argon in ice is quantified by 39Ar measurements on a long-stored ice core, showing the negligibility of contamination via this pathway for
storage times below several decades.
Finally, to make 39Ar, and also the gaseous dating tracers 81Kr and 85Kr, more conveniently applicable to ice dating, a new gas extraction system was designed, tested and applied to an Antarctic ice core. It enables the gas extraction from the otherwise unused waste line of a continuous flow analysis
Detection, quantification and 3D localization of anatomical changes with a helium-beam radiography system based on thin pixelated detectors
No full textThe clinical use of ion beams facilitates highly precise treatments of patients suffering from cancer. Utilizing the sharp peak in the depth-dose curve of protons and heavier ions, ion-beam radiotherapy treatments exhibit particularly localized dose distributions compared to conventional modalities. However, they are sensitive to changes such as variations in the patient’s anatomy which can occur over the treatment course of typically several weeks. These changes could be assessed prior to each treatment fraction by employing the ion beam itself to generate an image of the patient’s anatomy from beam’s eye view. Despite extensive efforts over the past decades, no ion-beam radiography system is routinely used in clinics yet.
Most ion-imaging prototypes require devices with a large material budget. In order to increase the practicability and foster clinical acceptance, a compact detection system was built from thin silicon pixel detectors. However, the system’s ability to precisely measure the clinically relevant quantity for heterogeneous objects is limited.
This thesis experimentally investigates the energy-painting approach eliminating this remaining principal limitation of the detection system based on thin detectors. Energy painting subsequently allows for the acquisition of a helium-beam radiograph of an anthropomorphic phantom, which is evaluated regarding its accuracy and compared to clinically used CT modalities based on X-rays. To extract the position of phantom modifications along the beam direction from single ion-beam projections, the 2.5D imaging method is investigated in simulations and experiments. Overall, the system’s capability to detect, quantify, and localize anatomical changes with high spatial resolution is demonstrated.
Consequently, this thesis presents fundamental contributions towards a future clinical application of the compact helium-beam radiography system for low-dose verification of the patient’s anatomy prior to ion-beam radiotherapy treatments
Modelling of long-term macrophage differentiation in vitro
No full textTissue macrophages lifespan can range from weeks to years. In pathological conditions such as cardiac inflammation or cancer, monocyte-derived macrophages undergo prolonged differentiation, and can adopt pathological programs. Most of previous research primarily focused on macrophage responses to short acting stimuli, while long-term macrophage differentiation and functional programming was
insufficiently explored. This thesis aimed to establish a reliable in vitro platform for investigating the prolonged differentiation process of macrophages derived from human monocytes. Specific aims included: 1) to establish conditions under which primary human monocyte-derived macrophages can differentiate over 24 days; 2) to analyze expression of genes that define the pro-inflammatory programming macrophages during their long-term differentiation; 3) to analyze scavenger receptors that define scavenging abilities of macrophages; 4) to examine the expression of metabolic regulators during long-term macrophage differentiation; 5) to examine α-SMA and Ki-67 as indicators of macrophage trans-differentiation and the transition from a proliferative to a non-proliferative phenotype. The final aim was to identify transcriptome differences between early and late stages of macrophage
differentiation. To establish the model, CD14+ human monocytes were differentiated under 2 conditions: M-CSF+IL-4, and M-CSF+IL-4+TGF-β. The combination of M-CSF + IL-4 resulted in the longest survival of macrophages in culture: up to 36 days. The expression of 22 genes, categorized into seven functional groups, was analyzed by RT-PCR during macrophage differentiation. These groups included cytokines, regulators of migration and recruitment, extracellular matrix modulators, chitinase-like proteins, scavenger receptors, metabolic regulators, and markers associated with the macrophage-to-myofibroblast transition (MMT). IL-4 alone and IL-4 in combination with TGF-β were used as stimuli in the long-term differentiation model. When IL-4 was combined with TGF-β, it had a more pronounced effect on the expression of genes linked to chronic inflammation in macrophages, compared to IL-4 alone. This combination strongly increased the expression of IL1RN, counteracting
the increase in pro-inflammatory IL1B, and also upregulated genes involved in inflammation and fibrosis, including FN1, CHI3L1, CHI3L2, OLR1, AKR1C3, SLC16A3, and ACTA2. Exposure to LPS, used to test the inflammatory reactivity of macrophages, triggered an increase in the expression of genes associated with inflammation, such as IL1B and IL1RN, migration and recruitment receptors such as CCR1 and CCR5, and metabolic reprogramming markers such as PFKFB3, an enhancer of aerobic glycolysis. Full transcriptome analysis was performed by NGS on macrophages differentiated under M-CSF+IL-4 stimulation for 6 and 36 days. The total number of upregulated genes on day 36 was 1356; the total number of
suppressed genes was 1003. Five most pronounced upregulated gene families on day 36 included aminoacyl-tRNA synthetases, lysosomal proteins, transporters, transcription factors, and aldehyde dehydrogenase. Five most suppressed gene families on day 36 were cholesterol biosynthesis pathway genes, solute carrier family, G protein-coupled receptors, integrins and adhesion molecules, and fatty acid and
lipid metabolism genes. These changes reflect the adaptation of macrophages to their environment, either by promoting inflammation and tissue damage or by modifying their metabolic state to maintain cellular functions under challenging conditions. This analysis allowed the identification of several biomarkers of long-term macrophage
differentiation in the context of IL-4 cytokine, including CHIT1, CHI3L1 and GDF15 that can be used for histological analysis in chronically inflamed or tumor tissues. Confocal microscopy analysis showed that macrophages stimulated with IL-4 at later stages of differentiation had the capacity to divide and to transdifferentiate into myofibroblast-like cells, which are key drivers of pathological remodeling and fibrosis. In summary, our model provides a useful tool to dissect macrophage behavior under prolonged stimulation. TGF-β, in particular, was found to drive expression patterns linked to chronic inflammatory states and pathologies such as fibrosis and cancer. This long-term culture model is crucial for studying the persistent inflammatory processes and functional alterations in macrophages over extended periods
Missbrauchskontrolle bei grenzüberschreitenden Umwandlungen nach der Mobilitätsrichtlinie mit besonderem Blick auf ihre Umsetzung im UmRUG
No full textScintillating Fiber-based Beam Profile Monitor for Particle Therapy
No full textA real-time beam profile monitor (BPM) has been developed for particle therapy and evaluated at the Heidelberg Ion Beam Therapy Center (HIT). The system is designed to provide real-time measurements of the ion beam position and width, thereby enabling precise radiation delivery in future clinical applications.
The BPM measures the one-dimensional projections of the beam using two layers of 250 µm scintillating fibers, which are read out by Hamamatsu S11865-64 photodiode arrays equipped with amplifier and integration circuits. The photodiode channels have dimensions of 0.7 mm (w) x 0.8 mm (h) with a pitch of 0.8 mm. The signals from the photodiodes are digitized on a custom PCB hosting voltage regulators and ADCs. Both the photodiodes and ADCs are controlled by a MAX 10 FPGA, which, in addition to signal control, performs ADC data packaging and beam profile reconstruction before transmitting the processed data via Ethernet to a PC equipped with dedicated data acquisition software.
This thesis investigates the signal characteristics of the SciFi-BPM, develops a tailored beam profile reconstruction algorithm, and implements it on the FPGA, enabling real-time determination of the beam position and width. Beam tests at HIT with protons and carbon ions demonstrated that the performance is sufficient for the majority of beam settings, specifically achieving an overall resolution better than 0.05 mm for the beam position and 0.12 mm for the beam width at high intensities
Model selection in the framework of multi-state models
No full textIn multi-state models based on high-dimensional data, effective modeling strategies are required to determine an optimal, ideally parsimonious model. In particular, linking covariate effects across transitions is needed to conduct joint variable selection. A useful technique to reduce model complexity is to address homogeneous covariate effects for distinct transitions. This approach is integrated to data-driven variable selection by extended regularization methods within multi-state model building. The fused sparse-group lasso (FSGL) penalized Cox-type regression is proposed in the framework of multi-state models combining the penalization concepts of pairwise differences of covariate effects along with transition-wise grouping. For optimization, the alternating direction method of multipliers (ADMM) algorithm is adapted to transition-specific hazards regression in the multi-state setting. In a simulation study and application to acute myeloid leukemia (AML) data, the algorithm’s ability to select a sparse model incorporating relevant transition-specific effects and similar cross-transition effects is evaluated. Settings in which the combined penalty is beneficial compared to global lasso regularization are investigated
Linguistically-Inspired Neural Coherence Modeling
No full textCoherence is an essential property of well-written text, making it easier to read and understand than a sequence of randomly arranged sentences. Assessing text coherence is valuable for many tasks. For example, it can be used to automatically score documents, reducing manual effort, or to provide feedback to students, helping them improve their writing quality. It can also serve as a reward model for training Large Language Models (LLMs) to generate more coherent and natural text. Given the importance of the task, many methods have been proposed for coherence modeling. Among these approaches, the dominant ones are neural network-based models due to their strength in representation learning and feature combination.
In linguistics, many factors contribute to achieving textual coherence. For example, text coherence can be achieved by describing the same set of entities or using discourse relations between sentences. However, existing work on neural coherence modeling focuses on using more powerful encoders or solely entity information, without a systematic analysis of the benefits of different linguistic features. In this thesis, we investigate the importance of entity- and relation-based patterns for coherence assessment and develop novel approaches to utilize these features individually or jointly.
We first investigate the benefits of entity-based patterns for coherence modeling. We analyze previous work that has leveraged entity patterns for coherence assessment. Then, we introduce a novel graph-based approach that captures the similarity of entity transition patterns between documents, rather than limiting the modeling of these patterns within a single document. We evaluate this approach on multiple benchmarks, and the results demonstrate that it outperforms various baselines.
Next, we examine the role of discourse relations in coherence modeling. Existing discourse parsers struggle with implicit discourse relation classification, limiting the use of discourse relations in coherence assessment. To address this, we propose a novel framework that jointly generates a connective between arguments and predicts discourse relations based on both the arguments and the generated connectives. Experiments show that our joint model achieves state-of-the-art performance on the PDTB 2.0, PDTB 3.0, and PCC datasets.
Beyond proposing a novel model for implicit discourse relation classification, we also investigate an unanswered question in the discourse processing community: why do relation classifiers trained on explicit examples (with connectives removed) perform poorly in real implicit scenarios? We identify label shift caused by the removal of connectives as a key factor contributing to this failure. To support this finding, we provide both manual analysis and corpus-level empirical evidence. Additionally, we propose two strategies to mitigate the impact of label shift.
Using the improved discourse parser, we identify discourse relations within documents and empirically demonstrate their correlation with textual coherence. Based on this observation, we develop a novel fusion model that integrates discourse relation-based features into a pre-trained model for coherence modeling.
Finally, we explore combining entity-based and discourse relation-based features for coherence modeling. This approach is motivated by the observation that writers typically employ multiple strategies simultaneously to ensure coherence. To this end, we design two methods to jointly model entities and discourse relations for coherence assessment. Experimental results demonstrate that both approaches significantly outperform models that use either features in isolation, highlighting the importance of considering both types of features simultaneously