11163 research outputs found
Sort by
Climate dynamics across scales in a mountain landscape: Macro- and microclimatic modelling in the Río Puelo basin, northern Patagonia
No full textMountain forests in northern Patagonia are exposed to rapid climatic changes, with increasing
temperatures, changing precipitation regimes, and more frequent and severe extreme events
threatening their current structure and resilience. Understanding these short- and long-term
dynamics requires integrating climate information across scales: from macroclimatic datasets
that inform broad biogeographic analyses to microclimatic measurements that capture the
conditions organisms directly experience. This dissertation addresses how macroclimatic and
microclimatic modelling can be combined to improve ecological assessments in mountainous
forest landscapes, focusing on the Río Puelo watershed of northern Patagonia.
Three empirical studies form the core of this work. The first compares two widely used high-
resolution climate datasets (CHELSA v2.1 and WorldClim v2.1), showing substantial
differences in precipitation estimates and associated bioclimatic classifications, with direct
implications for how forest distribution may be projected under future climate scenarios. The
second develops spatial models of microclimatic variability based on a dense network of in-
situ sensors combined with remote sensing, as well as weather station and reanalysis data.
Results highlight the strong and dynamic influence of vegetation structure and topographic
position on forest microclimates, affecting both near-ground and sub-canopy (2 m) temperature
and moisture conditions, and demonstrate the feasibility of modelling these patterns at 30 m
resolution. The third study examines microclimatic buffering during summer heat extremes,
revealing that forest structure and elevation consistently reduce maximum temperatures and
moderate diurnal warming and cooling rates, while slope orientation exerts weaker and more
variable effects. Importantly, buffering strength increases under extreme heat, underscoring
the temperature-sensitive character of microclimatic regulation.
Taken together, the findings show that the regions of highest ecological sensitivity are also
those where climate representation is most uncertain, and where fine scale vegetation and
topographic features exert the greatest influence on local conditions. By integrating
macroclimatic uncertainty with microclimatic evidence, this dissertation advances conceptual
and methodological links between climate and ecosystem research across scales. The results
have direct implications for modelling species distributions, understanding forest regeneration
dynamics, and informing ecosystem-based management strategies in the fire prone temperate
forests of northern Patagonia. Beyond the case study region, the work shows how combining
climate data across scales supports more robust evaluations of climate–ecosystem
relationships in mountainous terrain. It also provides a framework that can be applied to other
forested landscapes where microclimatic processes are critical for resilience. Finally, it
highlights the need to account for both uncertainty in global climate products and the buffering
capacity of local ecosystems when assessing biodiversity responses to climate change.2025-12-0
Learning the shape of neurons: Representation learning, generative modeling and hierarchical clustering
No full textThe remarkable diversity of neuronal morphologies has continued to fascinate researchers since Ramón y Cajal's pioneering work in 1911.
A neuron's dendritic shape determines the inputs it receives, and, consequently, the computations it can perform.
Therefore, in order to understand brain function, it is crucial to understand neuronal structure.
For this analysis, we need methods that can systematically describe the morphological features.
The advent of large-scale neuronal datasets has made it both necessary and feasible to use quantitative, data-driven approaches to analyze the neurons.
In this thesis, we focus on "learning the shape of neurons" by learning representations of neuronal morphologies and examining their organization in relation to cortical gradients, laminar structure, and expert-defined cell types.
To achieve this, we leverage a large-scale, serial-section electron microscopy dataset, containing more than 30,000 three-dimensional reconstructions of neuronal morphologies from the mouse early visual cortex.
First, we aim to learn meaningful, data-driven representations directly from the raw 3D morphology of neurons.
For this, we develop and refine methods for learning compact, low-dimensional feature embeddings -- "bar codes" -- that contain the neurons' structures.
These bar codes uniquely identify the neurons, thereby providing a meaningful similarity metric for comparing morphologies.
We evaluate the encoding performance of various supervised and unsupervised representation learning approaches by examining their latent spaces, and demonstrate alignment with established biological knowledge.
Second, we want to understand the principles underlying morphological diversity.
Therefore, we move beyond feature extraction, and introduce MorphOcc, a generative model of 3D surface shapes of neurons. The resulting learned embeddings not only capture morphological features with high detail but also support classification into expert-defined cell types.
Moreover, by interpolating within the embedding space, the generative model is able to synthesize novel instances of neuronal morphologies without supervision.
Third, given the latent spaces obtained in the previous steps, we raise the question as to how these embeddings are organized.
To this end, we propose t-NEB, a hierarchical clustering algorithm designed to uncover structure at multiple scales without the need to predefine the number of clusters.
As a proof of concept, we apply this method to multiple 2D synthetic datasets, as well as a real-world transcriptomic dataset.
We demonstrate that it can not only cluster accurately but also recover a biologically meaningful hierarchy of transcriptomic cell types.
These contributions establish a methodological foundation for modeling and understanding the diversity of neuronal shapes, offering a high potential to advance our knowledge about the neurons in the early visual system and beyond.
The code is available at https://github.com/ecker-lab/neuron-shape-learning.2025-12-1
Conditional Stability and Instability Estimates for Inverse Random Source Problems
No full textThis thesis is motivated by the study of passive imaging problems. Such problems can be characterised as scattering problems under uncontrolled incident waves. We focus specifically on inverse random source problems, a subclass of passive imaging problems. These are linear inverse problems where we aim to recover the random source from measurements of the scattered field on some surface enclosing the source region. In general, such problems lack uniqueness because of the existence of non-radiating sources. However, for centred Gaussian source processes, it can be shown that such inverse problems are uniquely solvable. Building on this uniqueness, we investigate the stability of these inverse problems. In particular, we prove that such problems satisfy Hölder-logarithmic stability estimates under Sobolev-type smoothness assumptions on the source power, depending on the wave number. For a fixed wave number, we also establish logarithmic instability estimates that only slightly differ in the order of the exponential ill-posedness compared to the conditional stability estimates. Furthermore, the used methodology enables us to derive convergence rates for the inverse problems for a variety of spectral regularisation methods under suitable parameter choice rules.
The first part of the thesis introduces conditional stability estimates within a generalised framework. In particular, we introduce a notion of local conditional stability, tailored to fit into the localised concept of variational source conditions. In the second part we focus on random processes in Hilbert spaces that serve as models for the random sources in the passive imaging. The third part presents the main contributions of this thesis, namely the conditional stability and instability estimates for the inverse random source problems. We study three different problems, and for each we use complex geometric optics solutions to establish variational source conditions, leading to Hölder-logarithmic stability and convergence rates. This part concludes with establishing logarithmic instability for each of the problems. Both logarithmic stability and instability rely on natural Sobolev-type smoothness assumptions on the source power of uncorrelated centred Gaussian source processes. Finally, we numerically investigate the convergence rates.2026-01-2
Phylogenetic and Evolutionary Analysis of Anthoshorea, Richetia, Shorea, and Rubroshorea (Dipterocarpaceae: Tribe Shoreae) Using Chloroplast and Nuclear Markers
No full textDiese Dissertation untersucht die phylogenetischen Beziehungen, die Merkmalsevolution und die molekulare Systematik ausgewählter Gattungen innerhalb der Familie Dipterocarpaceae, insbesondere jener, die dem Tribus Shoreae zugeordnet sind: Anthoshorea, Richetia, Shorea sensu stricto und Rubroshorea. Die Systematik der Dipterocarpaceae stellt für Taxonomen seit langem eine Herausforderung dar, da feine morphologische Unterschiede, uneindeutige Diversität und unregelmäßige Reproduktionszyklen die Klassifikation erschweren. Diese Studie verwendet ein auf mehreren Gensequenzen basierendes phylogenetisches Rahmenwerk, die Kartierung der Merkmalsevolution und baumbasierte Identifikationsmethoden, um die Taxonomie zu klären, evolutionäre Beziehungen zu bewerten und eine robuste Grundlage für zukünftige ökologische und molekulare Arbeiten innerhalb der Familie Dipterocarpaceae zu schaffen. Die Untersuchung beginnt mit einer Darstellung des historischen und biologischen Kontexts der Systematik von Dipterocarpaceae. Probleme in der Taxonomie, die sich aus einer Überbetonung vegetativer Merkmale ergeben, werden hervorgehoben. Solche Merkmale unterliegen häufig phänotypischer Plastizität, geografischer Variation und Hybridisierung, was eine verlässliche Artbestimmung erschwert. Die jüngste Neuklassifikation von Shorea sensu lato in spezifischere Gattungen erfordert eine weitere Validierung durch integrative Methoden. Daher verfolgt diese Studie drei Hauptziele: die Bewertung der Übereinstimmung zwischen chloroplasten- und kernbasierten Phylogenien, die Analyse der Integration von phylogenetischen, morphologischen und ökologischen Merkmalen sowie die Entwicklung einer baumbasierten Methode zur Artidentifikation. Diese Ziele tragen zum besseren Verständnis tropischer Baum-Systematik bei. Diese Forschung betont die Entwicklung der Taxonomie der Familie Dipterocarpaceae von einem morphologiebasierten hin zu einem modernen molekularen Ansatz. Sie liefert aktuelle Informationen zur Klassifikation von Shorea sensu lato und verwandten Arten, einschließlich morphologischer Merkmale wie Nebenblattform und Flügel des Fruchtkelchs. Die Studie zeigt auch Fortschritte im Einsatz von plastidären und nuklearen Markern in der Phylogenie mittels Bayesscher Inferenz und Maximum-Likelihood-Analysen. Diese Grundlage unterstützt den Übergang von phänotypischen zu kladistischen Ansätzen in der Systematik tropischer Bäume und identifiziert gleichzeitig bestehende Wissenslücken hinsichtlich Merkmalsevolution und Artabgrenzung auf molekularer Ebene. Die Analyse chloroplastischer und nuklearer phylogenetischer Bäume zeigte eine starke Übereinstimmung auf Gattungsebene für beide Topologien, wobei auf niedrigeren taxonomischen Ebenen Unterschiede auftraten. Beide Topologien wiesen gut unterstützte Kladen auf, insbesondere für die Gattungen des Tribus Shoreae, während die übrigen Triben der Dipterocarpoideae hauptsächlich als Referenz- oder Außengruppen einbezogen wurden. Die nukleare ITS-Sequenzierung ergab insgesamt ähnliche Beziehungen, jedoch wiesen mehrere Kladen eine geringe statistische Unterstützung auf, und bei einigen Arten wurden abweichende Platzierungen festgestellt, möglicherweise infolge von Hybridisierung oder unvollständiger Liniensortierung. Inkongruenzen wurden im Tanglegram deutlich, insbesondere bei Anthoshorea bracteolata, Hopea odorata und Hopea myrtifolia, die in den beiden Topologien unterschiedliche Positionen einnahmen. Ein Beispiel dafür ist die Probe Rubroshorea leprosula KR5454, die im Chloroplastenbaum mit Rubroshorea acuminata gruppiert wurde, im Kernbaum jedoch mit Rubroshorea leprosula, während zwei Proben von Rubroshorea singkawang (KR0865, KR0876) im ITS-Baum mit Rubroshorea dasyphylla gruppiert wurden. Diese Ergebnisse unterstreichen die Bedeutung der Kombination von chloroplasten- und kernbasierten Markern für eine genauere und stabilere Phylogenie, insbesondere bei Shoreae, die eine komplexe Evolutionsgeschichte aufweist. Fünf Schlüsselmerkmale wurden auf den Konsensusbaum abgebildet, um zu untersuchen, wie sich diese Merkmale parallel zur kombinierten Phylogenie aus fünf Markern (chloroplasten- und kernbasiert) entwickelt haben. Die untersuchten Merkmale waren Höhenverbreitung, Nebenblattform, Blattform, Form der Blütenknospe und Typ des Fruchtkelchflügels, die Werte für jede Art wurden aus der Literatur entnommen. Die Analyse ergab, dass Nebenblätter und Kelchmorphologie starke phylogenetische Signale zeigten, insbesondere innerhalb von Anthoshorea und Rubroshorea. Im Gegensatz dazu waren Blattform und Höhenverbreitung über die Phylogenie verteilt, was darauf hindeutet, dass diese Merkmale stärker durch Umweltfaktoren als durch gemeinsame Abstammung beeinflusst werden. Die Erkenntnis, dass nicht alle Merkmale das gleiche phylogenetische Gewicht tragen, liefert wichtige Hinweise für zukünftige taxonomische Schlüssel und merkmalsbasierte ökologische Studien. Sie stützt auch die Ansicht, dass morphologische Daten durch phylogenetische Belege validiert werden sollten, bevor sie in der Taxonomie verwendet werden. Auf Grundlage des Konsensusbaums wurde ein baumbasiertes Identifikationssystem entwickelt, das als Instrument zur molekularen Diagnostik und zur Abgrenzung von Arten dienen kann. Die Phylogenie, basierend auf fünf molekularen Markern, bestätigte die Monophylie aller vier Gattungen mit hoher posteriorer Wahrscheinlichkeit. Die meisten Arten wurden im Allgemeinen in deutlich abgegrenzte Cluster aufgelöst; jedoch zeigten mehrere Individuen eine artübergreifende Gruppierung. Zum Beispiel gruppierte Rubroshorea acuminata KR0753 mit Rubroshorea parvifolia subsp. parvifolia KR0508, und Rubroshorea acuminata KR0754 mit R. parvifolia subsp. parvifolia KR4953. Ebenso gruppierte Rubroshorea pinanga 41a mit Rubroshorea macrophylla 45c, während Rubroshorea macrophylla 45b mit Rubroshorea pinanga (41c, 49c) gruppiert wurde. Diese Muster deuten auf unvollständige Liniensortierung, mögliche Hybridisierung oder eine mögliche Fehlbestimmung von Feldproben hin. Arten innerhalb von Anthoshorea und Richetia zeigten eine klare Trennung in ihren jeweiligen Kladen. Shorea sensu stricto, dessen phylogenetische Klassifikation in der Vergangenheit problematisch war, bildet nun eine gut unterstützte Klade mit zuverlässiger Artplatzierung. Darüber hinaus wurden mittlere paarweise Distanzen zwischen den Gattungen, gattungsspezifische Nukleotidsubstitutionen sowie artspezifische diagnostische Einzelnukleotid-Polymorphismen (SNPs) identifiziert, die für diagnostische Zwecke nützlich sind. Diese Ergebnisse sind für den Naturschutz, die Wiederaufforstung und forstliche Forensik von großer Bedeutung, insbesondere zur Bestimmung von Sämlingen, frischen Proben und Holzmaterial. Die Ergebnisse dieser Arbeit eröffnen eine breitere Perspektive auf die Dynamik von Taxonomie und Evolution innerhalb der Familie Dipterocarpaceae. Die erfolgreiche Integration molekularer Daten mit morphologischen und ökologischen Merkmalen verstärkt das Potenzial der Multi-Locus-Phylogenetik zur Auflösung taxonomischer Unsicherheiten bei tropischen Bäumen. Diese Studie bestätigt die jüngsten taxonomischen Revisionen auf Gattungsebene und demonstriert die Anwendbarkeit baumbasierter Identifikationsmethoden in komplexen Systemen. Zudem bietet sie einen Weg zur Anwendung evolutionsbiologischer Rahmenwerke im Naturschutz und Ressourcenmanagement, insbesondere durch die Identifikation von Linien mit hoher evolutionärer Eigenständigkeit und begrenzter geografischer Verbreitung. Methodisch zeigt diese Arbeit, wie sequenzbasierte Werkzeuge und morphologische Daten gemeinsam analysiert werden können, um genauere und aussagekräftigere Klassifikationen zu erzielen. Sie unterstreicht auch die Bedeutung modellbasierter Ansätze wie der Bayesschen Inferenz, die phylogenetische Beziehungen mit hoher Sicherheit auflösen. Im Vergleich zu früheren Studien mit begrenzter Probenzahl und wenigen Loci stellt diese Forschung einen erweiterten Multi-Locus-Datensatz bereit, der als Referenz für zukünftige Arbeiten dienen kann. Es ist jedoch zu beachten, dass phylogenetische Analysen auf der Grundlage von genomweiten RADseq-Daten (z. B. Heckenhauer et al., 2018) eine noch höhere Auflösung liefern und daher eine wichtige Ergänzung für künftige Studien darstellen. Gleichzeitig wird der Bedarf an detaillierteren genomischen Daten und erweiterten geografischen Stichproben betont. Diese Arbeit hat auch über den akademischen Bereich hinaus praktische Relevanz. In der Naturschutzbiologie gewinnt die Identifikation und Priorisierung von Arten anhand phylogenetischer Merkmale zunehmend an Bedeutung, insbesondere in Biodiversitäts-Hotspots, die von Entwaldung und Klimawandel bedroht sind. In der Forstwirtschaft und Holzzertifizierung kann ein baumbasiertes Identifikationssystem, das auf gut aufgelösten Phylogenien basiert, die Artverifizierung verbessern und zur Vermeidung illegaler Holzernte beitragen. In der Wiederherstellungsökologie kann das Verständnis evolutionärer Beziehungen zwischen Arten die Herkunftsstrategien für Pflanzmaterial informieren und zur Erhaltung genetischer Vielfalt beitragen. Insgesamt leistet diese Dissertation einen bedeutenden Beitrag zur Systematik tropischer Pflanzen, indem sie die Taxonomie der Gattungen innerhalb des Tribus Shoreae verfeinert, Muster der Merkmalsevolution aufzeigt und den Wert der Multi-Locus-Phylogenetik für die Auflösung von Artenbeziehungen demonstriert. Sie schlägt eine Brücke zwischen traditioneller morphologischer Taxonomie und molekularen Daten und schafft eine starke Grundlage für zukünftige integrative Studien. Mit der zunehmenden Verfügbarkeit von Sequenzierungstechnologien und phylogenomischen Ansätzen bietet der in dieser Studie entwickelte Rahmen eine wertvolle Orientierung für Biodiversitätsdokumentation, Naturschutzplanung und nachhaltige Forstwirtschaft in Südostasien.This thesis investigates the phylogenetic relationships, trait evolution, and molecular systematics of selected genera within the Dipterocarpaceae family, specifically those classified under the tribe Shoreae: Anthoshorea, Richetia, Shorea sensu stricto, and Rubroshorea. Taxonomists have long faced challenges in the systematics of Dipterocarpaceae due to subtle morphological distinctions, ambiguous diversity, and irregular reproductive cycles. This study applies a multilocus phylogenetic framework, character evolution mapping, and tree-based species identification to clarify taxonomy, assess evolutionary relationships, and establish a robust foundation for future ecological and molecular work within the Dipterocarpaceae family. The investigation begins by outlining the historical and biological context of Dipterocarpaceae systematics. Taxonomic issues arising from an over-reliance on vegetative characters are highlighted. These characters are often influenced by phenotypic plasticity, geographic variation, and hybridization, which make species identification unreliable and unstable. The recent reclassification of Shorea sensu lato into more specific genera requires further validation through integrative methods. Therefore, this study was undertaken with three main objectives: to evaluate the comparison between chloroplast and nuclear phylogenies, to analyze the integration of phylogenetic analysis with morphological and ecological traits, and to develop a species identification approach through tree-based methods. These aims contribute to the broader understanding of tropical tree systematics. This research emphasizes the taxonomic evolution of the Dipterocarpaceae family, transitioning from a morphology-based framework to one grounded in modern molecular approaches. It provides updated information on how Shorea sensu lato and related species are classified, including morphological traits such as stipule shape and fruit calyx wings. The study also highlights the advancement in the use of plastid and nuclear markers in phylogenetics through Bayesian Inference and Maximum Likelihood analyses. This foundation supports the shift from phenetic to cladistic approaches in the systematics of tropical trees, while also identifying gaps in our understanding of character evolution and species delimitation from a molecular perspective. The analysis of chloroplast and nuclear phylogenetic trees revealed strong congruence at the genus level for both topologies, with variations observed at lower taxonomic levels. Both the chloroplast- and nuclear-based topologies showed well-supported clades, particularly for the genera within the tribe Shoreae, while the other tribes of Dipterocarpoideae were mainly included as reference or outgroups. The nuclear ITS recovered broadly similar relationships, but several clades showed low statistical support, and some differences in species placements were observed, possibly due to hybridization or incomplete lineage sorting. Incongruence was highlighted in the tanglegram, showing that species such as Anthoshorea bracteolata, Hopea odorata, and Hopea myrtifolia did not occupy the same positions in both the chloroplast and nuclear topologies. For example, Rubroshorea leprosula accession KR5454 grouped with Rubroshorea acuminata in the chloroplast tree but with R. leprosula in the nuclear tree, while two accessions of Rubroshorea singkawang (KR0865, KR0876) clustered with Rubroshorea dasyphylla in the ITS tree. These findings emphasize the importance of combining chloroplast and nuclear markers to achieve a more accurate and stable phylogeny, especially for the tribe Shoreae, which has a complex evolutionary history. Five key characters were used for mapping onto the consensus tree to examine how traits have evolved alongside the combined phylogeny generated from five markers using both chloroplast and nuclear datasets. The five main characters examined were elevation range, stipule shape, leaf morphology, flower bud shape, and the type of fruit calyx wing, with values for each species extracted from the literature. From the analysis, it was found that stipule and calyx morphology exhibited strong phylogenetic signals, particularly within the Anthoshorea and Rubroshorea groups. In contrast, leaf shape and elevation range were scattered across the phylogeny, suggesting that these traits may be more influenced by environmental pressures than inherited from a common ancestor. The finding that not all traits carry the same phylogenetic weight offers valuable insight for future studies in taxonomic keys and trait-based ecology. It also supports the view that trait data should be validated with phylogenetic evidence before relying on it for taxonomic classification. With strong support from the consensus tree, a tree-based identification system was developed due to its potential as a tool for molecular diagnostics and species delimitation. The phylogeny based on five molecular markers resolved all four genera as monophyletic with high posterior probability support. Most species were generally resolved into distinct clusters; however, several accessions showed cross-species grouping. For example, Rubroshorea acuminata KR0753 clustered with Rubroshorea parvifolia subsp. parvifolia KR0508, and Rubroshorea acuminata KR0754 grouped with Rubroshorea parvifolia subsp. parvifolia KR4953. Similarly, Rubroshorea pinanga 41a clustered with Rubroshorea macrophylla 45c, while Rubroshorea macrophylla 45b grouped with Rubroshorea pinanga (41c, 49c). These patterns indicate incomplete lineage sorting, potential hybridization, or possible misidentification of field samples. Species within Anthoshorea and Richetia showed clear separation, forming monophyletic groups within their respective clades. Meanwhile, Shorea sensu stricto, which previously had a problematic phylogenetic classification history, now exhibits a well-supported clade with reliable species placement, supported by high posterior probabilities. In addition, mean pairwise distances between genera, genus-specific nucleotide substitutions, and additionally species-diagnostic Single Nucleotide Polymorphisms (SNPs) were identified, offering diagnostic utility. These findings are highly applicable for conservation, restoration, and forestry forensics, especially in cases requiring identification of seedlings, fresh specimens, and timber samples. The findings of this thesis offer a broader perspective on the dynamics of taxonomy and evolution within the Dipterocarpaceae family. The successful integration of molecular data with morphological and ecological traits further strengthens the potential of multilocus phylogenetics to resolve taxonomic uncertainties in tropical tree systems. This study confirms the latest genus-level taxonomic revisions and demonstrates the practicality of tree-based identification methods in complex systems. Moreover, it provides a pathway for applying evolutionary frameworks to conservation and natural resource management, particularly by identifying lineages with high evolutionary distinctiveness and limited geographic distribution. Methodologically, this thesis illustrates how sequence-based tools and morphological data can be jointly analyzed to produce more accurate and informative classifications. It also highlights the importance of model-based approaches such as Bayesian Inference, which resolve phylogenetic relationships with high confidence. Compared to previous studies with limited sampling and loci, this research contributes an expanded multilocus dataset that can serve as a reference for future work. However, it should be noted that phylogenetic analyses based on genome-wide RADseq data (Heckenhauer et al., 2018) provide even higher resolution and thus represent an important complement for forthcoming studies. At the same time, it emphasizes the need for more detailed genomic data and expanded geographic sampling. This work has practical relevance beyond the academic realm. In conservation biology, identifying and prioritizing species based on phylogenetic characteristics is increasingly important, especially in biodiversity hotspots facing deforestation and climate change. In forestry and timber certification, a tree-based identification system grounded in well-resolved phylogenies can enhance species verification and help prevent illegal logging. In restoration ecology, understanding evolutionary relationships among species can inform population sourcing strategies and help maintain genetic diversity. Overall, this thesis makes a significant contribution to tropical plant systematics by refining the taxonomy of genera within the tribe Shoreae, elucidating patterns of trait evolution, and demonstrating the value of multilocus phylogenetics for resolving species relationships. It bridges the gap between traditional morphological taxonomy and molecular data and lays a strong foundation for future integrative studies. With the growing accessibility of sequencing technologies and phylogenomic approaches, the framework developed in this study will guide biodiversity documentation, conservation planning, and sustainable forest management across Southeast Asia.2026-02-0
Evolution of insect neural cell types revealed by single cell sequencing
No full textThe insect central nervous system, composed of the ventral nerve cord and brain, develops from neural stem cells (neuroblasts), which are largely conserved among insects. According to Drosophila studies, the brain develops from neuroblasts that eventually generate neural and glial cells that inherit the molecular identities of their progenitors. We know that the number of neuroblasts differs among insect species, but it is unclear to what extent their identities have evolved to produce the striking differences in size, structure, and complexity of an organ as diverse as the brain. I aimed to compare the neuroblast of the red flour beetle (Tribolium castaneum) and Drosophila to reveal their genetic makeup and degree of conservation and divergence using single-cell sequencing.
Chapter I presents a method to dissociate cells and FACS tGFP-marked neuroblasts from Tribolium embryos. A dataset of Drosophila neuroblast single nuclei sequencing has been previously reported by Veloso and Zinzen (Veloso and Zinzen, 2022). However, the dissociation and sorting methods used there were not suitable for Tribolium. So, I established a mechanical dissociation protocol to produce high-quality cell suspensions from a transgenic Tribolium line where the embryonic neuroblasts are marked with a green fluorescent protein (tGFP). These suspensions were pipelined to carry out cell sorting, and the resulting cells were suitable for performing single-cell RNA sequencing with 10X Genomics Chromium technology.
In chapter II, I address gene orthology between Drosophila and Tribolium. Comparing gene expressions among organisms requires a reliable assignment of orthologous genes. However, the existing assignments, as documented in iBeetle-Base, turned out to lack precision. Therefore, we produced a comparable gene annotation between Drosophila and Tribolium. Together with a colleague, we tackled this problem by using the reference protein predictions from Flybase and iBeetle-Base to compute phylogenetic orthology inferences using the OrthoFinder platform. In addition, we employed the eggNOG 6.0 database and manual phylogenetic tree analyses to enhance and assess the reliability of our results. We identified more than 9,000 orthologous genes between Drosophila and Tribolium that I used for annotating the Tribolium single-cell datasets. This set of genes will be a valuable resource for ongoing fly-beetle comparisons from other labs. We deposited this more reliable orthology assignment in iBeetle-Base for public interactive exploration or download.
Chapter III presents the main findings of this work. The goal was to use single-cell sequencing to comprehensively identify genes involved in the specification of Tribolium neuroblasts. By comparing these results to corresponding Drosophila data, I aimed to gain initial insights into the degree of conservation versus divergence in the genetic control of neuroblast number and identity. I produced single-cell sequencing datasets of the Tribolium neuroblasts and confirmed their identity by verifying the expression of neuroblast gene markers. To extract the brain neuroblasts from the dataset, I removed neuroblasts expressing Hox genes, which are known to be expressed in the ventral nerve cord (VNC), while the brain is primarily a Hox-free region. To comprehensively identify transcription factors that might play important roles in the brain compared to VNC neuroblasts, I performed a comparable differential expression analysis independently in Drosophila and Tribolium datasets. I identified numerous genes already known to be expressed and play significant roles anteriorly in Drosophila and Tribolium, including Optix, toy, tll, erm, fd102C, Rx, and oc, which confirmed the validity of my analysis. Interestingly, I also identified several genes not previously known to play roles in neuroblast development, such as Hmx, CG15696, CG32532, dmrt99B, fd59A, disco, Fer1, and TfAP-2, and confirmed their specific anterior expression using in situ hybridizations. Taken together, I present a comparative single-cell study of the neuroblasts of Tribolium and Drosophila, where I identified a comprehensive set of transcription factors expressed in brain neuroblasts that are likely involved in brain neuroblast determination in insects. Notably, I also discovered genes that appear to have divergent expression patterns in Drosophila and Tribolium and may contribute to differences in brain development. These results pave the way for future studies such as understanding the developmental role of the newly identified neuroblast transcription factors and determining their differing functions in these species.2026-01-2
Evaluation der sonographisch bestimmten Gallenblasenwanddicke als nicht-invasiver Marker für das Vorliegen eines Leberparenchymumbaus
No full textLiver cirrhosis is the final stage of various chronic liver diseases. Causes include chronic alcohol consumption, untreated chronic viral or autoimmune hepatitis, metabolic disorders, and medication, alongside cases with unclear etiology. Due to the high mortality of liver cirrhosis, early detection of liver parenchymal changes is critical for enabling targeted interventions that can slow down or even halt progression towards cirrhosis in some chronic liver diseases. While histological examination of liver tissue remains the gold standard for the detection of parenchymal changes, its invasive nature and potential for lethal complications have encouraged the development of non-invasive diagnostic techniques, including transient elastography and shear wave elastography. Additionally, laboratory-derived scoring systems provide a reliable non-invasive means of estimating the degree of parenchymal changes.
In this study, we evaluated whether gallbladder wall thickening is associated with sonographic and histological signs of liver parenchymal changes or cirrhosis. Furthermore, we examined whether gallbladder wall thickening correlates with liver pathology scores such as FIB-4, APRI, MELD, and Child-Pugh.
We conducted a retrospective monocentric cross-sectional evaluation of 505 patients with chronic hepatitis and suspected or confirmed liver cirrhosis. Of these, 145 patients had a histological biopsy result within two years before or after an abdominal ultrasound examination. Patients were divided into subgroups with and without sonographic signs of liver parenchymal changes. Additionally, patients with available histological results were categorized as having confirmed liver cirrhosis (Stage 4 according to Desmet and Scheuer) or not having cirrhosis (Stages 0 to 3 according to Desmet and Scheuer), to compare them regarding sonographically measured gallbladder wall thickness, laboratory parameters, and derived scores.
We identified an association between measured gallbladder wall thickness and the presence of sonographic liver parenchymal changes or histological liver cirrhosis. For the first time, a threshold value of gallbladder wall thickness for the detection of sonographic or histological liver parenchymal changes was calculated. In other studies, a cutoff value for relevant gallbladder wall thickening between 3 and 4 mm was established. In contrast, using ROC curve analysis and the Youden index, we calculated a threshold value for the presence of liver parenchymal changes. A gallbladder wall thickness of ≥3.3 mm was associated with sonographic signs of liver parenchymal changes. Furthermore, a gallbladder wall thickness of ≥3.3 mm was associated with histologically confirmed liver cirrhosis in Stage 4 according to Desmet and Scheuer. Patients with sonographic signs of liver parenchymal changes had significantly more often a histologically detectable cirrhosis in Stage 4 according to Desmet and Scheuer. Furthermore, we established a positive correlation between gallbladder wall thickness and the APRI and FIB-4 scores. Both scores are validated, non-invasive, and simple to calculate, and can predict the likelihood of liver cirrhosis without the need for histology.
By combining measurements of gallbladder wall thickness ≥3.325 mm with APRI >2 or FIB-4 score >3.325, a better discrimination of patient groups with sonographic and histological liver parenchymal changes was achieved compared to considering individual parameters. The accuracy of gallbladder wall thickness in detecting liver parenchymal changes or liver cirrhosis was higher in patients with sonographic signs of liver parenchymal changes than in patients with histologically confirmed liver cirrhosis.
We also found correlations between gallbladder wall thickness and liver and spleen sizes, gallbladder width, portal vein diameter, and other biochemical parameters. Additionally, positive correlations between gallbladder wall thickness and the MELD score and Child-Pugh stage were demonstrated.
Overall, the results suggest that sonographic measurement of gallbladder wall thickness in combination with other non-invasive parameters provides a useful contribution to the assessment of liver parenchymal changes. Accordingly, routine sonographic measurement of gallbladder wall thickness and further calculations of scores like APRI, FIB-4, MELD, and Child-Pugh stage could be a useful measure for early detection of liver parenchymal changes in patients with chronic liver diseases. This could help avoid severe complications from liver cirrhosis and provide time to initiate appropriate interventions.2025-02-1
Molecular assembly of auditory ribbon synapses in mice
No full textAuditorische Bändersynapsen sind spezialisierte Strukturen in den inneren Haarzellen (IHCs), die eine schnelle Neurotransmitterfreisetzung ermöglichen und somit für eine präzise Signalübertragung im auditorischen System essentiell sind. Derzeit fehlen jedoch Einblicke in die entwicklungsbedingte Bildung und Reifung dieser spezialisierten Synapsen. Diese methodische Doktorarbeit untersucht daher drei Ansätze zur Erforschung der Assemblierung auditorischer Ribbon-Strukturen: Zunächst analysierten wir die Verteilungsmuster der mRNA relevanter Proteine mittels RNAscope. Dabei modifizierten wir das Protokoll, sodass wir einerseits die Zellkörper mithilfe von Immunhistochemie abgrenzen und andererseits ganze Zellen durch „floating samples“ betrachten konnten. Anschließend versuchten wir, die subzelluläre Lokalisation der Translation von Ribbon-Proteinen mit einem Puromycin-basierten Proximity Ligation Assay (PLA) nachzuweisen. Obwohl wir die Aufnahme von Puromycin in die Zellen erfolgreich etablieren und optimieren konnten, erwies sich die Spezifität des PLAs im intakten Gewebe als unzureichend, was die Anwendbarkeit dieser Methode in diesem Kontext einschränkte. Schließlich verwendeten wir isotopisch markierte Aminosäuren und nanoscale secondary ion mass spectrometry („NanoSIMS“), um deren Aufnahme in die synaptischen Bänder zu untersuchen. Hier konnten wir Anzeichen einer erhöhten Aufnahme zeigen, was auf einen gesteigerten Proteinumsatz während der Entstehung der Ribbons hindeuten könnte. Wenn gleich RNAscope und NanoSIMS sich als vielversprechende Werkzeuge zur Untersuchung der Reifung von auditorischen Bändersynapsen darstellten, sind weitere Untersuchung mit größeren Stichprobengrößen notwendig, um diese Ergebnisse zu untermauern. Letztlich bildet diese Arbeit die Basis zur Erweiterung unseres Verständnisses der molekularen Mechanismen, die die Entwicklung der Ribbon-Synapsen steuern, und ebnet den Weg für fundamental neue, wertvolle Einblicke in die auditorische Verarbeitung.Auditory ribbon synapses are specialized structures in the inner hair cells (IHCs) that enable rapid neurotransmitter release, making them essential for precise signal transmission in the auditory system. However, insights into the developmental formation and maturation of these specialized synapses are still lacking. Investigating their composition provides crucial insights into their role in auditory development. This methodological doctoral thesis explores three approaches to studying the assembly of auditory ribbon structures.
First, we analyzed the distribution patterns of mRNA of relevant proteins at the mRNA level using RNAscope. In this process, we modified the protocol so that, on one hand, we could outline cell bodies using immunohistochemistry, and on the other hand, examine whole cells through "floating samples." Next, we attempted to detect the in vivo translation of ribbon proteins using a puromycin-based Proximity Ligation Assay (PLA). While we successfully optimized puromycin uptake into the cells, the specificity of the PLA proved insufficient, limiting the applicability of this method in this context. Finally, we used isotopically labeled amino acids and NanoSIMS imaging to identify uptake in ribbon-associated regions. Here, we were able to show signs of increased uptake, which might indicate enhanced protein turnover during ribbon maturation. While RNAscope and NanoSIMS appeared as promising tools for studying ribbon formation, further validation with larger sample sizes is necessary to support these results. Ultimately, this research enhances our understanding of the molecular mechanisms driving ribbon synapse development and thereby providing valuable insights into auditory processing.2025-02-1
Analyse von Klettersportler*innen
No full textIm Hinblick auf die zunehmende Beliebtheit des Klettersports wurden in einer prospektiven klinischen Studie eine Kohorte von 51 Kletter*innen aus dem Breitensport, welche in mittleren Schwierigkeitsgraden kletterten, mit einer Vergleichskohorte von 49 nicht- kletternden Athlet*innen im Hinblick auf verschiedene physiologische sowie psychologische Parameter statistisch untersucht.
Die Basisdaten der Proband*innen wurden über einen Fragebogen erfasst. Für die Griffkraftmessung verschiedener Griffformationen der dominanten und nicht-dominanten Seite wurde das sogenannte Leonardo Mechanograph® GF Pinch Dynamometer verwendet. Eine Vermessung der oberen Extremität erfolgte an verschiedenen Messpunkten mittels Maßband. Es wurden verschiedene Umfänge der Arme und Hand- sowie Armspannweite vermessen. Sowohl Griffkraftwerte als auch Messwerte der oberen Extremität wurden BMI- gewichtet betrachtet. Die psychologischen Parameter wurden über das deutschsprachige Big Five Inventar (BFI-2) untersucht. Die Risikobereitschaft wurde über einen Fragebogen in Anlehnung an die Risikobereitschaftskurzskala analysiert. Verletzungen der Kletter*innen wurden in einem Fragebogen erfasst. Die statistische Auswertung umfasste Mittelwerte, Standardabweichungen, die Signifikanz der Unterschiede zwischen den Kohorten und die Effektstärke.
Die beiden Kohorten unterschieden sich in ihrer Altersstruktur und ihrem BMI nicht signifikant voneinander. Die Griffkraft der Kletter*innen war sowohl auf der dominanten Seite als auch auf der nicht-dominanten Seite statistisch signifikant höher als die der Athlet*innen. Unterarmumfänge der Kletter*innen waren vor allem dann statistisch signifikant höher als die der Athlet*innen, wenn sie in Relation zum Body-Mass-Index oder in Relation zum Oberarmumfang betrachtet werden. In den psychologischen Merkmalen wiesen Kletter*innen lediglich in der Gewissenhaftigkeit statistisch signifikant höhere Werte auf. Die Kletter*innen verhielten sich in ihrer subjektiven Risikoeinschätzung sehr risikobewusst. Insgesamt wurden von den Kletter*innen mehr Verletzungen an der unteren als an der oberen Extremität angegeben.
Die Vorzüge dieser Arbeit sind die ausführliche Vermessung der Griffkraft, die Unterteilung in dominante und nicht-dominante Seite, die Anpassung der Messwerte auf den Body-Mass- Index der Proband*innen sowie die hohe Anzahl an Proband*innen zu nennen. In Folgestudien wäre eine ausführlichere Erfassung der Verletzungshistorie aller Proband*innen vorzuziehen. Mit den hier dargestellten Ergebnissen wurden Breitensport-Kletter*innen vielseitig charakterisiert. Die Ergebnisse können als Grundlage für weiterführende Studien in den Bereichen der Verletzungsprävention, Orthopädie und Unfallchirurgie sowie Sportwissenschaft dienen.With regard to the increasing popularity of climbing, a prospective clinical study was conducted. 51 amateur climbers, who climbed at medium levels of difficulty, were statistically compared with a cohort of 49 non-climbing athletes with regard to various physiological and psychological parameters.
The test subjects' baseline data were recorded using a questionnaire. The Leonardo Mechanograph® GF Pinch Dynamometer was used to measure the grip strength of various grip formations on the dominant and non-dominant side. The upper extremity was measured at several points using a tape measure. Both grip strength values and upper limb measurements were BMI-weighted. The psychological parameters were examined using the German Big Five Inventory (BFI-2). Risk appetite of the subjects was analyzed using a questionnaire. Injuries of the climbers were recorded. The statistical analysis included mean values, standard deviations, the significance of the differences between the cohorts and the effect size. The two cohorts did not differ significantly in their age structure and BMI. The grip strength of the climbers was found to be significantly higher than that of the non-climbing athletes on both the dominant side and the non-dominant side. The forearm circumferences of the climbers were significantly higher than those of the non-climbing athletes, especially when considered in relation to the body mass index or in relation to the upper arm circumference. In the psychological characteristics, climbers only showed significantly higher values for conscientiousness. The climbers were very risk-aware in their subjective risk assessment. Overall, the climbers reported more injuries to the lower than to the upper extremities.
The advantages of this study are the detailed measurements of grip strength, the division into dominant and non-dominant side, the adjustment of the measurements to the subjects' body mass index and the large number of subjects. In follow-up studies, a more detailed recording of the injury history of all test subjects would be preferable. The results presented here characterise amateur sports climbers in a number of ways. This study can serve as a basis for further studies in the fields of injury prevention, orthopaedics and trauma surgery as well as sports science.2025-02-2
Body region-specific germ and resistance spectrum of trauma surgery/orthopedic/plastic surgery patients in a maximum care clinic
No full textMuskuloskelettale Infektionen und bakterielle Arthritis sind häufige und gefürchtete Komplikationen bei Operationen. Um die Behandlung von postoperativen Infektionen zu optimieren, ist es sinnvoll, das Erregerspektrum der einzelnen Körperregionen zu kennen, um eine gezielte antibiotische Behandlung zu ermöglichen.
Es handelt sich um eine retrospektive Datenanalyse von 2015-2018 aus einem Krankenhaus der Maximalversorgung in der Unfallchirurgie, Orthopädie oder Plastischen Chirurgie. Insgesamt wurden 14.859 Operationen auf Keimnachweis untersucht und den entsprechenden Lokalisationen zugeordnet.
Insgesamt konnte ein individuelles Keimspektrum für die einzelnen Lokalisationen identifiziert werden. Einige bisher weniger bekannte Bakterien sollten als mögliche Differentialdiagnose bei postoperativen Infektionen erkannt und durch eine frühzeitige Materialentnahme nachgewiesen werden. In Zukunft sollte eine lokalisationsspezifische Antibiotikatherapie in Betracht gezogen werden.Musculoskeletal infections and bacterial arthritis are frequent and dreaded complications of surgery. To optimize the treatment of postoperative infections, it is useful to know the pathogen spectrum of the individual body regions to enable targeted antibiotic treatment.
This is a retrospective data analysis from 2015-2018 from a maximum care hospital in trauma surgery, orthopaedics or plastic surgery. A total of 14,859 operations were examined for germ detection and assigned to the corresponding localizations.
Overall, we were able to identify an individual germ spectrum for the individual localizations. Some previously less well-known bacteria should be recognized as a possible differential diagnosis in postoperative infections and should be detected by taking material at an early stage. In the future, localization-specific antibiotic therapy should be considered.2026-02-0
Relevance of methylated DNA fragments as a non-invasive biomarker in kidney failure
No full textEpigenetische Veränderungen, insbesondere die DNA-Methylierung, spielen eine zentrale Rolle bei der Progression akuter und chronischer Nierenerkrankungen. Die frühzeitige Identifikation von Risikopatienten und rechtzeitige Einleitung präventiver Maßnahmen sind entscheidend, um das Fortschreiten von Nierenerkrankungen und das Eintreten eines Nierenversagens zu verhindern. In dieser Arbeit wurden epigenetische Marker, insbesondere die Methylierung der Gene BMP7, EPO und RASAL1, untersucht, um deren Bedeutung als Biomarker für Nierenschäden und die Notwendigkeit einer Nierenersatztherapie zu evaluieren. Die Methylierung dieser Gene in peripheren Blutlymphozyten wurde durch methylierte DNA Immunpräzipitation und quantitative Realtime-PCR bestimmt. Die Methylierungsanalyse erfolgte an Blutproben zweier Patientenkohorten: Intensivpatienten mit akutem Nierenversagen und chronisch dialysepflichtigen Patienten. Es zeigten sich signifikante Korrelationen zwischen der Methylierung der Gene und verschiedenen Stadien der Nierenerkrankung: BMP7 war mit akutem Nierenversagen assoziiert, während EPO die chronischen Prozesse widerspiegelte. RASAL1, als Kontrollgen, korrelierte mit Fibroseprozessen, jedoch nicht spezifisch mit den akuten oder chronischen Stadien. In der Kohorte mit akutem Nierenversagen wurde ein Cut-Off-Wert für BMP7 identifiziert, der eine rechtzeitige Dialyseindikation ermöglichte. In der Kohorte mit chronischer Niereninsuffizienz bestätigte sich die Verlässlichkeit der Messung der Methylierung in Leukozyten und zeigte keinen signifikanten Einfluss der Dialyse auf die Methylierungswerte. Die Ergebnisse dieser Studie belegen die potenzielle klinische Anwendung der Methylierungsanalyse von BMP7, EPO und RASAL1 zur Vorhersage des Krankheitsverlaufs und zur frühen Einleitung therapeutischer Maßnahmen, insbesondere der Dialysebehandlung.Epigenetic modifications, particularly DNA methylation, play a central role in the progression of acute and chronic kidney diseases. Early identification of at-risk patients and timely initiation of preventive measures are crucial to halting disease progression and preventing kidney failure. This study investigated epigenetic markers, specifically the methylation of the BMP7, EPO, and RASAL1 genes, to evaluate their potential as biomarkers for kidney damage and the need for renal replacement therapy. DNA methylation levels in peripheral blood lymphocytes were determined using methylated DNA immunoprecipitation combined with quantitative real-time PCR. Methylation analysis was conducted on blood samples from two patient cohorts: critically ill patients with acute kidney injury and patients with chronic kidney disease undergoing dialysis. Significant correlations were observed between gene methylation patterns and various stages of kidney disease: BMP7 was associated with acute kidney injury, while EPO reflected chronic disease processes. RASAL1, used as a control gene, was linked to fibrotic processes but showed no specific association with either acute or chronic stages. In the acute kidney injury cohort, a cutoff value for BMP7 methylation was identified, enabling a timely indication for dialysis. In the chronic kidney disease cohort, methylation measurements in leukocytes proved reliable and were unaffected by dialysis. These findings demonstrate the potential clinical utility of BMP7, EPO, and RASAL1 methylation analysis for predicting disease progression and facilitating early therapeutic interventions, particularly dialysis initiation.2026-01-2