Sondersammlungen

Permanent URI for this communityhttps://hohpublica.uni-hohenheim.de/handle/123456789/7011

Browse

Now showing 1 - 20 of 24

Comparing hops and malt price transmission in the beer value chain: evidence from Germany
(2025) Hess, Sebastian; Bublik, Nikolas
The German beer value chain has received limited attention so far, despite the country’s central role in global beer production. This study investigates the price dynamics of its two key inputs—hops and malt—using monthly price data from 2015 to 2024 based on a unique dataset from a German hops cooperative. While contract farming is common for both raw materials, malt is traded via private firms, whereas hops are marketed almost exclusively through farmer-owned cooperatives. A vector error correction model (VECM) is estimated, incorporating structural break dummies identified through Bai–Perron tests, followed by forecast error variance decomposition (FEVD) and impulse response function (IRF) analysis. The results show that hop prices are largely self-driven and adjust more quickly to deviations from equilibrium than malt or beer prices. While malt and beer exhibit stronger interdependencies, the hop sector displays greater price stability. The findings further reveal that the malting sector responded significantly to the recent energy price crisis in Germany, whereas the hop sector did not.
A diamine oxidase from Glutamicibacter halophytocola for the degradation of histamine and tyramine in foods
(2025) Kettner, Lucas; Freund, Alexander; Bechtel, Anna; Costa-Catala, Judit; Fischer, Lutz
A novel diamine oxidase (DAO) was discovered in the bacterium Glutamicibacter halophytocola (DAO-GH). The gene of DAO-GH was integrated into the genome of the yeast Komagataella phaffii and recombinantly produced under control of the methanol-inducible AOX1 promoter in a bioreactor cultivation. A high DAO activity of 70.2 ± 5.2 µkat/Lculture (5.25 ± 0.22 µkat/gprotein) was yielded after 90 h of cultivation. The DAO-GH was partially purified by the polyethyleneimine precipitation of nucleic acids, fractionated ammonium sulfate precipitation and hydrophobic interaction chromatography, resulting in a specific DAO activity of 19.7 µkat/gProtein. The DAO-GH was then biochemically investigated regarding its potential for histamine and tyramine degradation in fermented foods and the human small intestine. Interestingly, the DAO-GH showed activity even at a low pH of 5 and low temperature of 6 °C. Both histamine and tyramine were effectively degraded and DAO-GH showed especially very high affinity towards tyramine (Km of 0.009 mM). The DAO-GH was shown to be capable of degrading around 20% of the initially applied histamine in tuna paste (pH 5.6) at 5 °C within 24 h and completely degraded the histamine in a simulated intestinal fluid within 1.5 h in bioconversion experiments. The DAO-GH was spray-dried for the production of a storable enzyme preparation. Only around 17% of activity were lost in this process and the DAO-GH remained stable at room temperature for at least 3 months. The discovery of this DAO with its very advantageous biochemical properties allows the preparation of histamine-reduced or -free fermented foods by a simple enzymatic treatment or the treatment of histamine intolerance symptoms as a dietary supplement or medicine.
Does a smarter ChatGPT become more utilitarian?
(2026) Pfeffer, Jürgen; Krügel, Sebastian; Uhl, Matthias; Pfeffer, Jürgen; Technical University of Munich, TUM School of Social Sciences and Technology, Munich, Germany; Krügel, Sebastian; Faculty of Business, Economics and Social Sciences, University of Hohenheim, Stuttgart, Germany; Uhl, Matthias; Faculty of Business, Economics and Social Sciences, University of Hohenheim, Stuttgart, Germany
Hundreds of millions of users interact with large language models (LLMs) regularly to get advice on all aspects of life. The increase in LLMs’ logical capabilities might be accompanied by unintended side effects with ethical implications. Focusing on recent model developments of ChatGPT, we can show clear evidence for a systematic shift in ethical stances that accompanied a leap in the models’ logical capabilities. Specifically, as ChatGPT’s capacity grows, it tends to give decisively more utilitarian answers to the two most famous dilemmas in ethics. Given the documented impact that LLMs have on users, we call for a research focus on the prevalence and dominance of ethical theories in LLMs as well as their potential shift over time. Moreover, our findings highlight the need for continuous monitoring and transparent public reporting of LLMs’ moral reasoning to ensure their informed and responsible use.
Effect of liquefaction temperature and enzymatic treatment on bioethanol production from mixed waste baked products
(2025) Almuhammad, Mervat; Kölling, Ralf; Einfalt, Daniel
This study investigates the effect of different liquefaction temperatures (50–70 °C) and four commercial enzyme formulations on glucose release and subsequent ethanol yield, using mixed waste baked products as a substrate. Among the enzymes tested, Amylase GA 500 proved to be superior in the hydrolysis of starch at lower temperatures (50 °C and 55°C). At higher liquefaction temperatures (65 °C and 70°C) all four enzyme preparations showed comparable activity. The highest glucose concentration (205.7 g/L) and the highest ethanol yield (92 g/L) were achieved with Amylase GA 500 at 65 °C. Its superior performance is attributed to the synergistic activity of α-amylase and glucoamylase, which facilitates efficient starch hydrolysis. Crucially, we discovered that the liquefaction temperature profoundly affects fermentation speed independently of the initial glucose concentration or the enzyme preparation used for starch hydrolysis. This novel mechanistic insight suggests that higher temperature treatment either makes an additional factor crucial for yeast fermentation available or depletes/destroys an inhibitor present in the complex waste bakery product matrix. These findings highlight the critical role of temperature and enzyme formulation in optimizing bioethanol production from bakery waste, supporting the development of more sustainable and efficient waste-to-biofuel processes.
Enabling adaptive food monitoring through sampling rate adaptation for efficient, reliable critical event detection
(2025) Jox, Dana; Schweizer, Pia; Henrichs, Elia; Krupitzer, Christian; Jox, Dana; Department of Food Informatics and Computational Science Hub, University of Hohenheim, 70599 Stuttgart, Germany; Schweizer, Pia; Department of Food Informatics and Computational Science Hub, University of Hohenheim, 70599 Stuttgart, Germany; Niu, Jianwei; Department of Food Informatics and Computational Science Hub, University of Hohenheim, 70599 Stuttgart, Germany; Niu, Jianwei
Monitoring systems are essential in many fields, such as food production, storage, and supply, to collect information about applications or their environments to enable decision-making. However, these systems generate massive amounts of data that require substantial processing. To improve data analysis efficiency and reduce data collectors’ energy demand, adaptive monitoring is a promising approach to reduce the gathered data while ensuring the monitoring of critical events. Adaptive monitoring is a system’s ability to adjust its monitoring activity during runtime in response to internal and external changes. This work investigates the application of adaptive monitoring—especially, the adaptation of the sensor sampling rate—in dynamic and unstable environments. This work evaluates 11 distinct approaches, based on threshold determination, statistical analysis techniques, and optimization methods, encompassing 33 customized implementations, regarding their data reduction extent and identification of critical events. Furthermore, analyses of Shannon’s entropy and the oscillation behavior allow for estimating the efficiency of the adaptation algorithms. The results demonstrate the applicability of adaptive monitoring in food storage environments, such as cold storage rooms and transportation containers, but also reveal differences in the approaches’ performance. Generally, some approaches achieve high observation accuracies while significantly reducing the data collected by adapting efficiently.
Endogenous task allocation and intrafirm bargaining: a note
(2025) Marczak, Martyna; Beissinger, Thomas
We develop a model that incorporates task-based production into a matching model with intrafirm wage bargaining. Unlike in existing task-based models, the representative firm derives the optimal task allocation as a function of capital and labor, rather than relative factor prices. Embedding this mechanism in a model with strategic employment choice, we show how the properties of task-level technology affect the extent of overhiring.
EvaMol : A python tool for evaluating molecules in hit-to-lead optimization
(2025) Herzog, Anna-Maria; Steuber, Julia; Fritz, Günter
This Python script was developed as a tool in structure-based drug discovery processes, such as fragment-to-lead-optimization, where a large number of variants of an initially identified hit molecule have to be evaluated and ranked in silico. The tool facilitates the identification and selection of follow-up drug candidates with improved predicted pharmacokinetic and binding properties. These candidates can derive from different procedures like similarity search or systematic chemical modifications. The initial hit data are provided either as coordinates of the protein-molecule complex obtained experimentally or by in silico methods such as docking making the script a versatile tool adaptable to variable workflows.
From coffee waste to wastewater treatment: optimization of hydrothermal carbonization and H₃PO₄ activation for Cr(VI) adsorption
(2026) Piccoli Miranda de Freitas, Caroline; De Freitas Batista, Gabriel; Dalmolin da Silva, Mariele; Checa Gomez, Manuel; Arauzo, Pablo J.; França da Cunha, Fernando; Kruse, Andrea
Spent coffee grounds (SCG) are an abundant agro-industrial waste, and their valorization as activated carbon (AC) offers a sustainable approach for wastewater treatment and heavy-metal remediation. However, the high energy demand of SCG activation limits large-scale application. Hydrothermal carbonization (HTC) reduces energy consumption and enhances material properties. This study evaluated the performance of activated carbon (AC) derived from SCG via HTC, followed by H₃PO₄ activation for Cr(VI) removal, and compared it with non-activated carbon obtained by HTC and pyrolysis. The results highlight the effect of chemical activation on enhancing surface area, porosity, and adsorption efficiency. The predicted optimal IN was 1624.7 mg·g⁻¹, closely matching the experimental value of 1640.1 ± 15.5 mg·g⁻¹, achieved at 426 °C, 92 min, and a hydrochar-to-H₃PO₄ ratio of 1:1.6. The optimized AC exhibited a maximum adsorption capacity (Qₑ) of 33 ± 1.1 mg·g⁻¹ and 99.4 ± 0.1 % Cr(VI) removal under pH 2, 25 mg·L⁻¹ initial concentration, and 2 g·L⁻¹ adsorbent dose. In contrast, the non-activated carbon presented a lower iodine number (1411 ± 70 mg·g⁻¹) and inferior adsorption performance, confirming the key role of H₃PO₄ activation in improving surface reactivity and adsorption sites. Chemical activation proved essential for improving Cr(VI) adsorption, with the H₃PO₄-AC exhibiting the highest capacity. These results demonstrate the potential of SCG-derived AC as a low-cost adsorbent for heavy-metal-rich industrial effluents, supporting circular economy strategies.
Iron partitioning and photosynthetic performance in Cannabis sativa L. reveal limitations of nanoscale zero-valent iron as a fertilizer
(2025) Büser, Christian; Hartung, Jens; Deurin, Lukas; Graeff-Hönninger, Simone
Iron (Fe) is the fourth most abundant element in the Earth’s crust but remains the third most limiting nutrient for crop productivity due to its low solubility in most soils. The emergence of nanotechnology has introduced nanoscale zero-valent iron (nZVI) as a potential Fe fertilizer with high surface reactivity and improved bioavailability. However, its comparative efficacy relative to conventional chelated Fe sources remains poorly understood. This study investigated Fe partitioning, photosynthetic efficiency, biomass accumulation, and cannabinoid synthesis in Cannabis sativa L. grown hydroponically under Fe-EDTA, nZVI, or Fe-deficient (-Fe) treatments. Total Fe concentrations were markedly reduced in -Fe plants compared with both Fe-EDTA and nZVI treatments. Despite similar root Fe contents between Fe-EDTA and nZVI, only Fe-EDTA facilitated efficient translocation to shoots, while nZVI-derived Fe predominantly accumulated in roots. Consequently, nZVI-treated plants exhibited intermediate photosynthetic performance and water-use efficiency—lower than Fe-EDTA but significantly higher than -Fe. Although Fe translocation differed substantially, inflorescence biomass and cannabinoid yield were comparable between nZVI and Fe-EDTA treatments, both exceeding those of -Fe plants. These results suggest that yield reductions under Fe deficiency arise not solely from Fe scarcity but also from the metabolic costs of Strategy I Fe acquisition, which are partially circumvented by root Fe availability from nZVI. Overall, Fe-EDTA demonstrated superior nutrient use efficiency, whereas nZVI partially alleviated Fe deficiency and revealed distinctive interactions between nanomaterials and plant Fe physiology. This study advances understanding of nZVI as an alternative Fe source in C. sativa and provides new insights into nanoparticle–plant nutrient dynamics.
Leveraging regionally sourced natural fibers for coreless filament winding in sustainable construction
(2026) Rossa, Alina J.; Bozó, Daniel Christopher; Mindermann, Pascal; Gresser, Götz T.; Schuenemann, Franziska
This study investigates the potential of regionally sourced natural fibers (NF) as sustainable feedstocks for coreless filament winding (CFW), an innovative and resource-efficient composite fabrication method for construction. As the sector moves towards bio-based materials, flax has emerged as a benchmark due to its compatibility with CFW and favorable mechanical properties. However, supply chain vulnerabilities and price volatility highlight the need for regional alternatives to flax fibers. By assessing hemp, miscanthus, lavender, nettle and wheat straw cultivated in Baden-Württemberg (BW), this paper links crop sustainability and fiber suitability with the broader aim of strengthening local biomass utilization within the bioeconomy. A multi-stage evaluation framework compares environmental, economic, social, and technical indicators against flax, focusing on biomass availability, input intensity, ecosystem services, land-use competition, and fiber properties. The results identify hemp as the most promising alternative due to its high fiber output, low cultivation inputs, beneficial ecosystem services and mechanical performance. Nettle offers strong mechanical performance but faces agronomic limitations, while lavender presents niche potential under integrated production systems. Miscanthus and wheat straw, while unsuitable for direct CFW due to fiber morphology, may serve as supplementary feedstocks through modified processing. These findings underscore the role of local biomass streams in creating robust regional value chains, contributing to the resilience and sustainability of the bioeconomy. While the focus of this study is on BW, the framework can be transferred to other regions and may serve as a practical tool for policy, industry, and research seeking to advance circular approaches in bio-based construction.
Miscanthus‐derived products for material applications: can they contribute to greenhouse gas emission mitigation?
(2025) Lask, Jan; Weik, Jan; Kiesel, Andreas; Lewandowski, Iris; Wagner, Moritz; Lask, Jan; Institute of Crop Science, University of Hohenheim, Stuttgart, Germany; Weik, Jan; Institute of Crop Science, University of Hohenheim, Stuttgart, Germany; Kiesel, Andreas; Institute of Crop Science, University of Hohenheim, Stuttgart, Germany; Lewandowski, Iris; Institute of Crop Science, University of Hohenheim, Stuttgart, Germany; Wagner, Moritz; Institute of Applied Ecology, Geisenheim University, Geisenheim, Germany
Miscanthus is a particularly promising lignocellulosic biomass as it can also grow under marginal conditions and can be used for a wide range of products including energy and material applications. The latter, including applications in the construction, textile, chemical, or agricultural sector, is becoming increasingly relevant today. In general, it is hypothesised that biobased products are advantageous in terms of their greenhouse gas (GHG) performance when compared to conventional—in particular fossil—alternatives. To investigate this, the life cycle assessment methodology is typically applied. However, assessments are subject to uncertainty and variability due to assumptions and methodological choices. Given the increasing interest in miscanthus‐derived material applications, this study aims to draw more general conclusions about their GHG performance and relative mitigation potential. This should support a better understanding of their contribution to climate change mitigation objectives and guide the selection of promising products or product groups. A systematic review of peer‐reviewed literature was conducted. In total, 20 studies reporting on 188 comparisons of the GHG performance of miscanthus‐derived and alternative products were assessed. Most comparisons indicated potential GHG mitigation through miscanthus‐derived products, with the majority ranging between 20% and 100% savings. Key parameters defining the relative performance include the selection of the reference product, consideration of soil carbon changes, changes in product and process design, as well as the incorporation of indirect Land Use Change (iLUC) impacts. Overall, we conclude that miscanthus‐derived material applications have the potential to contribute to GHG emission mitigation if iLUC effects are minimised. Given the limited availability of agricultural land, miscanthus‐derived products with high absolute GHG mitigation potential per unit of biomass used and long product lifetime are preferable. For future development, potential environmental trade‐offs need to be monitored.
Model-based process design for surfactin production with Bacillus subtilis
(2025) Hiller, Eric; Off, Manuel; Dittmann, Holger; Perino, Elvio Henrique Benatto; Lilge, Lars; Hausmann, Rudolf; Hiller, Eric; Department of Bioprocess Engineering, Institute of Food Science and Biotechnology, University of Hohenheim, Stuttgart, Germany; Off, Manuel; Department of Bioprocess Engineering, Institute of Food Science and Biotechnology, University of Hohenheim, Stuttgart, Germany; Dittmann, Holger; Department of Bioprocess Engineering, Institute of Food Science and Biotechnology, University of Hohenheim, Stuttgart, Germany; Perino, Elvio Henrique Benatto; Department of Bioprocess Engineering, Institute of Food Science and Biotechnology, University of Hohenheim, Stuttgart, Germany; Lilge, Lars; Department of Bioprocess Engineering, Institute of Food Science and Biotechnology, University of Hohenheim, Stuttgart, Germany; Hausmann, Rudolf; Department of Bioprocess Engineering, Institute of Food Science and Biotechnology, University of Hohenheim, Stuttgart, Germany
Bacillus subtilis is one of the most important production organisms in industrial biotechnology. However, there is still limited knowledge about the kinetics of fed-batch processes in bioreactors, as well as a lack of biological performance indicators, such as production yields, particularly regarding their variation over time. Understanding these kinetics and changes is crucial for optimizing the productivity in fed-batch processes. Fed-batch bioreactor cultures of Bacillus subtilis BMV9 in high cell density processes for surfactin production have been characterized with a kinetic model composed of first-order ordinary differential equations, describing the time course of biomass, substrate, surfactin and acetate. This model contributes to understanding critical restrictions and the knowledge gained was used to design and implement a model-based process. The model integrates biomass growth based on Monod kinetics, substrate consumption, surfactin synthesis and formation of the by-product acetate. After the model was parameterized for B. subtilis BMV9 using 12 different fed-batch bioreactor experiments, the kinetic model was able to accurately describe biomass accumulation, substrate consumption, product formation rates and, to some extent, the overflow metabolism involving acetate. Based on this, the kinetic model was used for a process design, in which the batch was omitted, which led to a product titre of 46.33 g/L and a space–time-yield of 2.11 g/(L*h) was achieved. The kinetic model developed in this study enables the description of the time course of biomass growth, substrate consumption and product formation and thus significantly improves process understanding. The computation of process parameters, which are not analytically accessible at any time, could be realized. A sensitivity analysis identified the maximum specific growth rate, substrate-related maintenance and the maximum acetate formation rate as key parameters influencing model outputs.
Motivational framing strategies in health care information security training: randomized controlled trial
(2025) Keller, Thomas; Warwas, Julia Isabella; Klein, Julia; Henkenjohann, Richard; Trenz, Manuel; Thanh-Nam Trang, Simon
Background: Information security is a critical challenge in the digital age, especially for hospitals, which are prime targets for cyberattacks due to the monetary worth of sensitive medical data. Given the distinctive security risks faced by health care professionals, tailored Security Education, Training, and Awareness (SETA) programs are needed to increase both their ability and willingness to integrate security practices into their workflows. Objective: This study investigates the effectiveness of a video-based security training, which was customized for hospital settings and enriched with motivational framing strategies to build information security skills among health care professionals. The training stands out from conventional interventions in this context, particularly by incorporating a dual-motive model to differentiate between self- and other-oriented goals as stimuli for skill acquisition. The appeal to the professional values of responsible health care work, whether absent or present, facilitates a nuanced examination of differential framing effects on training outcomes. Methods: A randomized controlled trial was conducted with 130 health care professionals from 3 German university hospitals. Participants within 2 intervention groups received either a self-oriented framing (focused on personal data protection) or an other-oriented framing (focused on patient data protection) at the beginning of a security training video. A control group watched the same video without any framing. Skill assessments using situational judgment tests before and after the training served to evaluate skill growth in all 3 groups. Results: Members of the other-oriented intervention group, who were motivated to protect patients, exhibited the highest increase in security skills (ΔM=+1.13, 95% CI 0.82-1.45), outperforming both the self-oriented intervention group (ΔM=+0.55, 95% CI 0.24-0.86; P=.04) and the control group (ΔM=+0.40, 95% CI 0.10-0.70; P=.004). Conversely, the self-oriented framing of the training content, which placed emphasis on personal privacy, did not yield significantly greater improvements in security skills over the control group (mean difference=+0.15, 95% CI –0.69 to 0.38; P>.99). Further exploratory analyses suggest that the other-oriented framing was particularly impactful among participants who often interact with patients personally, indicating that a higher frequency of direct patient contact may increase receptiveness to this framing strategy. Conclusions: This study underscores the importance of aligning SETA programs with the professional values of target groups, in addition to adapting these programs to specific contexts of professional action. In the investigated hospital setting, a motivational framing that resonates with health care professionals’ sense of responsibility for patient safety has proven to be effective in promoting skill growth. The findings offer a pragmatic pathway with a theoretical foundation for implementing beneficial motivational framing strategies in SETA programs within the health care sector.
The non-nutritive sweetener rebaudioside a enhances phage infectivity
(2025) Marongiu, Luigi; Brzozowska, Ewa; Brykała, Jan; Burkard, Markus; Schmidt, Herbert; Szermer-Olearnik, Bożena; Venturelli, Sascha
Non-nutritive sweeteners (NNS) are widely employed in foodstuffs. However, it has become increasingly evident that their consumption is associated with bacterial dysbiosis, which, in turn, is linked to several health conditions, including a higher risk of type 2 diabetes and cancer. Among the NNS, stevia, whose main component is rebaudioside A (rebA), is gaining popularity in the organic food market segment. While the effect of NNS on bacteria has been established, the impact of these sweeteners on bacterial viruses (phages) has been neglected, even though phages are crucial elements in maintaining microbial eubiosis. The present study sought to provide a proof-of-concept of the impact of NNS on phage infectivity by assessing the binding of rebA to phage proteins involved in the infection process of enteropathogenic bacteria, namely the fiber protein gp17 of Yersinia enterocolitica phage φYeO3-12 and the tubular baseplate protein gp31 of Klebsiella pneumoniae phage 32. We employed docking analysis and a panel of in vitro confirmatory tests (microscale thermophoresis, RedStarch ™ depolymerization, adsorption, and lysis rates). Docking analysis indicated that NNS can bind to both fiber and baseplate proteins. Confirmatory assays demonstrated that rebA can bind gp31 and that such binding increased the protein’s enzymatic activity. Moreover, the binding of rebA to gp17 resulted in a decrease in the adsorption rate of the recombinant protein to its host but increased the Yersinia bacteriolysis caused by the whole phage compared to unexposed controls. These results support the hypothesis that NNS can impair phage infectivity, albeit the resulting effect on the microbiome remains to be elucidated.
Nutrient–response modeling with a single and interpretable artificial neuron
(2025) Rodehutscord, Markus; Ahmadi, Hamed
Precise estimation of nutrient requirements and utilization efficiency is fundamental to nutritional sciences, yet it is mainly performed using classical nonlinear regression models. These models are interpretable but require careful selection of the functional form and initial parameter values. Flexible machine learning (ML) methods are seemingly disliked due to their perceived “black box” nature, which can obscure biological insight. A minimal and interpretable ML framework addresses this gap in nutrient–response modeling. The proposed approach uses a single artificial neuron with a hyperbolic tangent activation. Mathematically, this resembles a four-parameter sigmoidal function but with greater flexibility and distinct parameter definitions, allowing capture of the monotonic, saturating dynamics typical of essential nutrient responses. The method is enhanced with modern ML best practices, including data augmentation, Bayesian regularization, and bootstrap resampling, providing robust, uncertainty-quantified estimates of key nutritional metrics—such as asymptotic response, inflection point, and nutrient requirements—even from small datasets. Evaluations across 12 diverse datasets from poultry and fish studies, including amino acids and phosphorus, demonstrated that the single artificial neuron matches or exceeds the performance of classical models while providing full analytical transparency. The framework is implemented as a no-code graphical application, ‘NutriCurvist’, offering an easy-to-use alternative tool for nutrient-response modeling to support data-driven, precision nutrition.
Poultry perfection : comparison of computer vision models to detect and classify poultry products in a production setting
(2025) Einsiedel, Daniel; Vita, Marco; Jox, Dana; Dunnewind, Bertus; Meulendijks, Johan; Krupitzer, Christian
This study explores the use of computer vision, specifically object detection, for quality control in ready-to-eat meat products. We focused on a single process step, labeling products as “good” or “imperfect”. An “imperfect product” constitutes a product that deviates from the norm regarding shape, size, or color (having a hole, missing edges, dark particles, etc.). Imperfect does not mean the product is inedible or a risk to food safety, but it affects the overall product quality. Various object detectors, such as YOLO, including YOLO12, were compared using the mAP50-95 metric. Most models achieved mAP scores over 0.9, with YOLO12 reaching a peak score of 0.9359. The precision and recall curves indicated that the model learned the “imperfect product” class better, most likely due to its higher representation. This underscores the importance of a balanced dataset, which is challenging to achieve in real-world settings. The confusion matrix revealed false positives, suggesting that increasing dataset volume or hyperparameter tuning could help. However, increasing the dataset volume is usually the more difficult path since data acquisition and especially labeling are by far the most time-consuming steps of the whole process. Overall, current models can be applied to quality control tasks with some margin of error. Our experiments show that high-quality, consistently labeled datasets are potentially more important than the choice of the model for achieving good results. The applied hyperparameter tuning on the YOLO12 model did not outperform the default model in this case. Future work could involve training models on a multi-class dataset with hyperparameter optimization. A multi-class dataset could contain more specific classes than just “good” and “imperfect,” making trained models capable of actually predicting specific quality deviations.
Prüfverfahren für UHF-RFID-Ohrmarken bei Schweinen: Weiterentwicklung von Labor- und Praxistestmethoden
(2025) Förschner, Adriana Johanna Martha; Gallmann, Eva
Diese Dissertation untersucht Ultrahochfrequenz-RFID-Transponderohrmarken (UHF-RFID) für die elektronische Einzeltiererkennung bei Ferkeln und Schweinen unter Labor- und Praxisbedingungen. Angesichts der wachsenden Bedeutung von UHF-RFID in der Nutztierhaltung besteht ein hoher Bedarf an standardisierten, praxisnahen Prüfverfahren, da bislang existierende Laborstandards nur eingeschränkt mit realen Einsatzbedingungen übereinstimmen. Ziel der Arbeit war es daher, eine bestehende Testumgebung systematisch zu optimieren und ein anwendungsorientiertes Prüfverfahren für UHF-RFID-Ohrmarken zu entwickeln. Im ersten Teil wurden Laboruntersuchungen an einem statischen Prüfstand durchgeführt. Die teilweise Auskleidung des Prüfbereichs mit elektromagnetischen Absorbern reduzierte Sweet Spots und Dead Spots und führte zu stabileren Messergebnissen sowie einer verbesserten Wiederholbarkeit. Diese Effekte waren beim Prüffaktor Ohr geringer, dennoch wird der Einsatz von Absorbern für zukünftige Messungen empfohlen. Der zweite Teil befasste sich mit der Entwicklung eines flüssigkeitsbasierten Ohrphantoms, dessen dielektrische Eigenschaften an Schweineohrgewebe angepasst wurden. Das Phantom ermöglichte reproduzierbare Messungen, zeigte transpondertypspezifische Leistungsunterschiede und unterstrich die Bedeutung einer zielgerichteten Vertrimmung von Transpondern auf das Schweineohr. Der Praxisteil validierte UHF-RFID-Ferkelohrmarken in einem Praxisversuch (Ferkelaufzucht). Alle Transponder blieben über den gesamten Zeitraum lesbar; Unterschiede ergaben sich durch Lesetechnik, Gruppenverhalten und Ohrwachstum. Die Ergebnisse verdeutlichen, dass Laborprüfstände eine wichtige Vorauswahl ermöglichen, Praxisversuche jedoch weiterhin unerlässlich sind. Das entwickelte kombinierte Vorgehen bietet eine belastbare Grundlage für zukünftige Prüf- und Entwicklungsprozesse von UHF-RFID-Ohrmarken bei Schweinen.
Recombinant production of Paenibacillus wynnii β-galactosidase with Komagataella phaffii
(2024) Bechtel, Anna; Seitl, Ines; Pross, Eva; Hetzel, Frank; Keutgen, Mario; Fischer, Lutz
The β-galactosidase from Paenibacillus wynnii (β-gal-Pw) is a promising candidate for lactose hydrolysis in milk and dairy products, as it has a higher affinity for the substrate lactose (low KM value) compared to industrially used β-galactosidases and is not inhibited by the hydrolysis-generated product D-galactose. However, β-gal-Pw must firstly be produced cost-effectively for any potential industrial application. Accordingly, the yeast Komagataella phaffii was chosen to investigate its feasibility to recombinantly produce β-gal-Pw since it is approved for the regulated production of food enzymes. The aim of this study was to find the most suitable way to produce the β-gal-Pw in K. phaffii either extracellularly or intracellularly.ResultsFirstly, 11 different signal peptides were tested for extracellular production of β-gal-Pw by K. phaffii under the control of the constitutive GAP promoter. None of the signal peptides resulted in a secretion of β-gal-Pw, indicating problems within the secretory pathway of this enzyme. Therefore, intracellular β-gal-Pw production was investigated using the GAP or methanol-inducible AOX1 promoter. A four-fold higher volumetric β-galactosidase activity of 7537 ± 66 µkatoNPGal/Lculture was achieved by the K. phaffii clone 27 using the AOX1 promoter in fed-batch bioreactor cultivations, compared to the clone 5 using the GAP promoter. However, a two-fold higher specific productivity of 3.14 ± 0.05 µkatoNPGal/gDCW/h was achieved when using the GAP promoter for β-gal-Pw production compared to the AOX1 promoter. After partial purification, a β-gal-Pw enzyme preparation with a total β-galactosidase activity of 3082 ± 98 µkatoNPGal was obtained from 1 L of recombinant K. phaffii culture (using AOX1 promoter).ConclusionThis study showed that the β-gal-Pw was produced intracellularly by K. phaffii, but the secretion was not achieved with the signal peptides chosen. Nevertheless, a straightforward approach to improve the intracellular β-gal-Pw production with K. phaffii by using either the GAP or AOX1 promoter in bioreactor cultivations was demonstrated, offering insights into alternative production methods for this enzyme.
Selective coupling or genuine integration? Startups in the twin transformation
(2026) Kuckertz, Andreas; Hinderer, Sebastian
Startups actively shape innovation during the twin transformation to sustainability and digitalization. Our analysis of >21,000 ventures reveals that genuine twin-transformers, i.e., startups able to integrate digital and green technologies, are rare and primarily driven by software. While sustainable ventures readily adopt digital tools, digital ventures rarely embrace sustainability. This asymmetry reflects different configurations of complementarities between digital and sustainable technologies, which investor preferences reinforce. Eventually, this may lead to superficial coupling rather than deep integration. To address the issue, we offer targeted recommendations for practitioners, policymakers, and researchers aiming to foster scalable, hardware-based solutions critical for sustainability.
Structure formation in fruit preparations by fruit fermentates produced with exopolysaccharide-forming lactic acid bacteria
(2025) Festini, Silvan; Zipori, Dor; Wallisch, Marc; Weiss, Agnes; Neidhart, Sybille; Schmidt, Herbert; Jekle, Mario
Fruit preparations are intermediate food products that are primarily used in the dairy industry for the production of fruit yogurt or frozen desserts. Typically, they are stabilized by added hydrocolloids like pectins. The objective of this study was to investigate the potential replacement of conventional stabilizers by structure-forming fermentates produced by exopolysaccharides (EPS)-forming lactic acid bacteria (LAB). Peach puree was selected as fermentation matrix. Prior to 72 h of incubation, it was inoculated with either the heterofermentative LAB strain Levilactobacillus brevis TMW 1.2112 or the homofermentative LAB strain Pediococcus parvulus strain LTH 1110, both being known to produce EPS in form of β-D-glucan. The lyophilized fermentates were applied as stabilizers to produce strawberry fruit preparations. Flow curves, viscoelastic behaviour and shear stability were measured to investigate the effect of fermentate incorporation on the rheological properties of the products. A fermentatively induced effect was observed in terms of a 1.3-fold increase in viscosity of strawberry model fruit preparations with 10 % fermentate of Lv. brevis TMW 1.2112 compared to the addition of the same dose of fermentate blank. Further, increasing the fermentate blank dose from 10 % to 15 % resulted in a 2.4-fold viscosity increase of the model fruit preparations. High shear stability was found in all model strawberry fruit preparations. However, fermentation had no clear benefit in terms of viscoelastic behaviour and shear stability of the fruit preparations. Although the fermentatively induced thickening potential was limited, production of viscosity-increasing peach fermentate with minor changes in the sugar and amino acid profiles of the fruit proved to be feasible.

Browse

Browsing Sondersammlungen by Sustainable Development Goals "9"