Willkommen bei JLUpub

JLUpub ist das institutionelle Repositorium der Justus-Liebig-Universität.

JLUpub bietet Mitgliedern und Angehörigen der Universität die Möglichkeit neben wissenschaftlichen Dokumenten auch Forschungsdaten elektronisch zu veröffentlichen und dauerhaft zugänglich zu machen. Alle Veröffentlichungen erhalten einen Digital Object Identifier (DOI) und werden über nationale und internationale Bibliothekskataloge sowie Suchmaschinen nachgewiesen und auffindbar.

Photo by B. Zimmermann

Hauptbereiche in JLUpub

Wählen Sie einen Bereich, um dessen Inhalt anzusehen.

Gerade angezeigt 1 - 2 von 2

JLUdata
Forschungsdatenrepositorium
JLUdocs
Schriftenrepositorium

Neue Veröffentlichungen:

Chromatic and luminance processing during eye movements in human early visual cortex
(2025) Zhang, Yuan
Human vision is not a passive process. Rather, our visual perception of the world emerges from a dynamic and adaptive interplay between sensory input and motor action, whereby perception continually guides movements, and these movements in turn reshape perceptual experience. Eye movements are essential for acquiring clear visual input in natural viewing. Despite substantial progress in understanding how eye movements are generated, controlled, and functionally organized, fundamental questions remain about how early visual processing is modulated by natural viewing behaviors, and how different types of eye movements influence neural mechanisms encoding color and luminance. This dissertation addresses these questions by combining precise neurophysiological measurement techniques with carefully controlled behavioral paradigms. Our eyes move frequently to keep objects of interest projected onto the fovea for a clear image. Among these movements, saccades are rapid, ballistic shifts of gaze that move the eyes from one location to another, while smooth pursuit keeps a moving target within the foveal region through continuous, slow eye rotations that closely match the target’s speed and direction. When the gaze is relatively stable and directed at a single point, the eye is in a state of fixation. These three types of eye movements are fundamental for acquiring visual information from the environment. Visual scenes are encoded by the L, M, and S cones in the retina and then relayed via three pathways: the magnocellular pathway (L+M, luminance), the parvocellular pathway (L–M, red-green opponency), and the koniocellular pathway (S–[L+M], blue-yellow opponency). These pathways transmit information in parallel up to the primary visual cortex (V1), where their signals are subsequently processed through partially distinct yet interacting cortical circuits (for reviews, see Gegenfurtner, 2003; S. H. C. Hendry & Reid, 2000; Nassi & Callaway, 2009). Given that V1 is the first cortical site where luminance and chromatic signals converge and begin to interact, a key unresolved question is how these signals are modulated by different types of eye movements during natural viewing. To explore the neural mechanisms underlying these processes, this thesis employed steady-state visual evoked potentials (SSVEPs) to track neural responses in the early visual cortex. SSVEPs are brain oscillations elicited by periodic visual stimulation (Adrian & Matthews, 1934; for review, see Norcia et al., 2015), originating primarily from V1 (Di Russo et al., 2007; Müller et al., 1997). SSVEPs possess high temporal resolution and exhibit narrowband spectral responses locked precisely to the stimulation frequency, making them highly resistant to eye movement artifacts (e.g., J. Chen et al., 2017a, 2017b; J. Chen, Valsecchi, et al., 2019). Those properties make SSVEPs a reliable tool for studying visual processing in the human early visual cortex during eye movements. In Study 1, we investigated the effect of chromatic (L–M) adaptation during prolonged fixation. Results showed that SSVEP responses to chromatic stimuli progressively decrease as stimulation duration increased, following an exponential decay with a half-life of approximately 20 seconds. In contrast, responses to luminance stimuli did not show any systematic adaptation. After characterizing this sustained visual adaptation, Study 2 then investigated transient modulations of visual cortical responses induced by saccadic eye movements. Results demonstrated comparable saccadic suppression effects on SSVEP responses to both chromatic (L–M) and luminance stimuli. Further modeling of contrast response functions revealed that saccades selectively reduced response gain without altering contrast gain, suggesting that visual attenuation involves a multiplicative mechanism operating similarly within both the parvocellular and magnocellular pathways. To enhance the quality of SSVEP data, the third study evaluated various EEG referencing methods and introduced the Laplacian reference as an optimal strategy for signal derivation. Results showed that the Laplacian reference significantly improved the signal-to-noise ratio (SNR) and reliability of neural measurements, while also being straightforward to implement across different experimental settings. Taken together, these findings underscore the critical role of eye movements in modulating both luminance and chromatic signals within the early visual cortex. This work demonstrates that perception is dynamically shaped by the continuous interplay between sensory input and oculomotor behavior, offering new insights into how active vision operates under natural viewing conditions.
Können Plaqueindizes die tatsächliche Plaquemenge abbilden?
(2025) Eilert, Franziska
Für die zahnmedizinische Prävention spielen plaquereduzierende Maßnahmen eine zentrale Rolle. Um einschätzen zu können, ob Mundhygienemaßnahmen effektiv sind und inwieweit sie die mikrobielle Besiedlung der Zahnoberflächen reduzieren, kommen heute verschiedene Methoden der Plaquequantifizierung zum Einsatz. Die zwei gängigsten Verfahren sind die Plaqueplanimetrie und der Einsatz von Plaqueindizes, wobei sich Letztere in ihren Differenzierungsgraden teils deutlich unterscheiden. Folglich wird die vorhandene Plaquemenge indexabhängig unterschiedlich abgebildet, was wiederum einen Einfluss auf die Vergleichbarkeit zwischen Studienergebnissen hinsichtlich der ermittelten Effektgröße und statistischen Signifikanz nimmt. Bei der Planimetrie hingegen wird die vorhandene Plaque objektiv durch eine Computersoftware als prozentualer Anteil an der Gesamtoberfläche errechnet. Ziel der vorliegenden Arbeit war es herauszufinden, inwiefern verschiedene Indizes in der Lage sind, die planimetrisch gemessene, tatsächliche Plaquebedeckung, auch zu unterschiedlichen Zeitpunkten der Mundhygiene, darzustellen. Der Erkenntnisgewinn diente unter anderem der kritischen Auseinandersetzung mit der Aussagekraft von vorherigen klinischen Studien. Hierzu wurden Intraoralscans von 30 Probanden zu jeweils drei unterschiedlichen Zeitpunkten der Mundhygiene angefertigt. Der erste Termin repräsentierte dabei die habituelle Mundhygiene (T1). Beim zweiten Termin stellten sich die Probanden nach 72 Stunden ohne Mundhygiene vor (T2), wurden anschließend dazu aufgefordert ihre Zähne wie gewohnt zu putzen und unterzogen sich dann dem dritten Intraoralscan (T3). Vor jeder Datenerfassung wurden die vorhandenen mikrobiellen Beläge durch einen Plaquerevelator (Mira-2-Ton®) sichtbar gemacht. Anhand der ermittelten 3D-Scans wurden standardisierte Screenshots der Vestibulär- und Oralflächen der Ramfjordzähne erstellt und die Plaquemenge mithilfe eines Grids hinsichtlich vier verschiedener Indizes ausgewertet (T-QHI, RMNPI, RMNPI nach Bretz, marginaler Plaqueindex). Im Anschluss erfolgte die Messung der tatsächlichen Plaquebedeckung mithilfe der Planimetrie. Es konnte gezeigt werden, dass zwar alle Indizes Plaquemengenveränderungen widerspiegeln können, diese aber in einer deutlich abweichenden Effektgröße repräsentiert werden. Die Indexwerte fielen immer signifikant niedriger als die planimetrisch erhobenen Messergebnisse aus, zudem bestand ein nichtlinearer Zusammenhang. Insgesamt lieferte der T-QHI noch die zuverlässigsten Ergebnisse, kann aber auch nur eingeschränkt einen Anhalt auf die tatsächliche Plaquemenge geben. Schlussfolgernd ist es also nicht sicher möglich Effektgrößen mithilfe von Plaqueindizes auszudrücken.
Ein Schatz aus dem späten 16. Jahrhundert: ein Nürnberger Gesangbuch
(2025) Schneider, Olaf
Unraveling the role of heterotrophic feeding in coral tolerance to ocean warming and microplastic pollution
(2025) López Hernández, María Antonieta Panamá
Coral reefs harbor the highest biodiversity of all marine ecosystems and support the livelihoods of nearly 500 million people worldwide. Yet, they are increasingly threatened by global warming. Marine heatwaves disrupt the symbiosis between corals and Symbiodiniaceae algae, leading to coral bleaching. As these algae provide up to 90% of coral energy requirements, their loss results in energy deficits that compromise coral health, reducing growth and reproduction, increasing disease susceptibility, and often leading to mortality. Heterotrophic feeding, as a secondary pathway of energy acquisition in reef-building corals, can sustain up to 100% of their metabolic demands under stress, thereby enhancing survival and resilience to bleaching caused by ocean warming. Consequently, trophic plasticity, i.e., the ability to modulate trophic strategies in response to environmental change, has emerged as a key trait distinguishing potential “winners” under climate change. However, trophic plasticity varies among species and may be compromised by emerging stressors such as microplastic pollution, which can interfere with coral feeding. Despite the importance of heterotrophy in coral resilience, there are still knowledge gaps regarding the role of food type and food availability on coral physiology, and how these interact with pollutants such as microplastic, particularly in an ocean warming context. These uncertainties are further compounded by a lack of field-based data. This thesis addresses these knowledge gaps through two controlled laboratory experiments and the first in situ application of compound-specific isotope analysis of amino acids during a natural bleaching event. Five coral species (Galaxea fascicularis, Porites lobata, Stylophora pistillata, Ctenactis echinata, and Pocillopora verrucosa) were assessed for key physiological traits (photophysiology, growth, energy reserves, δ¹⁵N and δ¹3C), providing a comparative framework of heterotrophic strategies. Results indicated that complex diets enhanced the benefits of heterotrophy across coral species. Although these benefits varied among species when symbiotic, all bleached corals exhibited positive responses to more complex food sources. Physiological rates were consistently higher in symbiotic fragments compared to their bleached counterparts, with the magnitude of these differences increasing alongside the baseline productivity of the species, from G. fascicularis to P. lobata and S. pistillata. Food treatments did not affect respiration or photosynthetic rates, suggesting that growth gains were driven primarily by enhanced heterotrophic nutrient supply. In situ bleaching was associated with δ¹⁵N enrichment in the trophic amino acids alanine (ALA), glutamic acid (GLU), isoleucine (ISO), proline (PRO), and valine (VAL), suggesting alterations in nitrogen acquisition and processing under stress. The trophic position (TP) of the symbiotic host and symbionts of C. echinata was 1.3, consistent with a mixotrophic diet. In contrast, P. verrucosa exhibited TP values of 1.2 for the host and 1.0 for the symbionts, indicating a stronger reliance on autotrophy. Under bleaching, C. echinata maintained a TP of 1.4, reflecting a stable mixotrophic strategy, whereas P. verrucosa shifted to a TP of 1.5, suggesting an increased reliance I on heterotrophy. These findings suggest higher heterotrophic plasticity in P. verrucosa compared to C. echinata. Microplastics (MPs) exposure significantly decreased the energy reserves of P. verrucosa, although it increased photosynthesis and respiration. High food availability partially mitigated the loss of tissue energy content observed while maintaining photosynthesis and respiration rates comparable to control conditions. S. pistillata was not affected by MP exposure alone, but when combined with high feeding, photosynthesis decreased below that of the Control. When exposed to short-term heat stress, all corals bleached severely, however, both species bleached less in the MP treatment suggesting that MPs may also disrupt the relationship between energy balance and thermal resilience. Overall, this thesis provides evidence supporting the central role of heterotrophic feeding in modulating coral tolerance to the combined pressures of ocean warming and microplastic pollution, while emphasizing the need to integrate trophic plasticity and interspecific variability into future conservation and management strategies.
Predictive Modelling with Machine Learning in Plant Breeding
(2025) Heilmann, Philipp Georg
Genomic prediction, originally proposed as a solution to the limitations of marker-assisted selection for complex traits, has become the standard for estimating breeding values in both inbred and hybrid crops. While linear models such as GBLUP and RR-BLUP remain effective in many cases, especially when assuming an additive genetic architecture, recent years have seen a growing interest in applying machine learning (ML) methods to overcome some of their constraints, including their limited capacity to model non-additive effects and nonlinear interactions. This thesis explored the influence of three key aspects on the success of genomic prediction: The choice of input features, the statistical model used, and the target trait or crop.
In terms of input features, marker data was compared to minimalist parentage-based models, haplotype blocks, and features generated using autoencoders. It was shown that even simple ML models using parentage-based information can rival marker-based GBLUP under certain conditions, which holds potential for small breeding programs with large amounts of historical, but ungenotyped, records. At the same time, dimensionality reduction techniques, especially a novel haplotype-based autoencoder that was developed during this thesis, were introduced to compress genomic data while preserving prediction accuracy and successfully accelerated model training.
Concerning the model aspect, a variety of ML algorithms were benchmarked using different approaches for hyperparameter tuning. Although no single model outperformed others across all traits and crops, ensemble approaches typically performed better than the individual models they were based on. Support vector machines seemed to be relatively unstable when compared to other ML based algorithms, such as tree-based models.
Finally, results showed that the accuracy of the genomic predictions was strongly dependent on differences between traits, crops with different breeding schemes, and different populations. For hybrids, ML performed well when SCA was more important for determining the hybrid yield than GCA. Large differences were observed for different fungal diseases in wheat, while differences among methods for the same disease were relatively similar.
While ML has not yet provided a significant improvement over traditional methods in many scenarios, its flexibility and potential for multi-modal data integration remain promising. The development of plant breeding-specific model architectures, such as haplotype-based autoencoders, may represent a more promising path than the general application of standard ML models.