Knorr-Bremse (Germany)

Analyzing microbial colonies on agar plates is a critical task in microbiology, with applications spanning research, diagnostics, and industry. Despite advances in automated systems, challenges remain in accurately segmenting and clustering colonies due to variability in their appearance and distribution. To address these challenges, we present a fully automated modular system for colony segmentation and clustering, combining state-of-the-art deep learning for segmentation and machine learning for cluster count prediction and clustering. The main contribution of this work is the proposition, development and evaluation of a novel system, which achieved a V-measure of 0.532 under real-world conditions, improving to 0.727 with ideal segmentation and cluster counts, setting a new benchmark for microbiological analysis. At the core of the system we propose AgarNet, an Attention U-Net-based architecture combined with an EfficientNetB4 backbone, which achieved an F1-score of 0.906 for segmentation. We also introduce a new BRUKERCLUSTER dataset, which is one of the largest and most diverse annotated resources of its kind, featuring expert-annotated images from a controlled cultivation. By combining the dataset, robust segmentation, accurate cluster count prediction, and effective clustering, the proposed system delivers a scalable solution for advancing automated microbial colony analysis. These results establish a new benchmark for automated colony clustering in microbiology, and by releasing the unique BRUKERCLUSTER dataset, we provide a valuable tool for future advancements in automated colony analysis.

An increased awareness to health, environment and animal welfare has led to a growing public interest in vegan products, and the market for meat‐ and dairy‐free alternatives is growing rapidly. Characterization of flavor and aroma profiles is of particular interest during the development of vegan substitutes of non‐vegan products. Since a large part of flavor and aroma perception arises from volatile compounds, GC‐HRMS is often used for analysis. However, in non‐target studies conventional El (electron ionization) combined with NIST library searches suffers from an ambiguous identification of unknowns due to less specific fragmentation, missing molecular ion signals or absence in reference libraries. Presented will be a novel GC‐HRMS that uses simultaneously both chemical ionization (Cl) for the exact molecular mass assignment and El for NIST searchable fragment spectra in a single GC run. 10g of vegan cheese from New Roots AG as well as its supposed non‐vegan equivalent (Appenzeller Classic) were incubated for 10 mins in a headspace vial. Each cheese was sampled in triplicate at 60°C for 30 mins using a Restek Polyacrylate SPME fiber (BGB Analytik, Switzerland). The GC‐HRMS consists of a high resolution TOF analyzer operating simultaneously a standard 70 eV El source and a medium pressure Cl source with an automated selection of different Cl reagents ions (NH 4 + , N 2 H + , H 3 O + It enables the adjustment of reactant selectivity and the degree of fragmentation for various compound classes of interest. Using both pieces of information, El and Cl, in a single run enables a highly confident compound identification and quantification in complex matrices. Between the two types of cheese a clear difference between common flavor compounds such as butanoic acid and 2/3‐ methyl butanoic acid can be observed. Several peaks of interest for both the vegan and the non‐vegan equivalent were selected at different retention times to evaluate profile differences. Statistical tools like volcano plots revealed potential markers for both samples. The value of additional Cl information due to the selection of different Cl reagents ions results in increased compound identification yields and certainties. Especially when El library hits are only accounted as „fair” with low corresponding probability, the additional chromatographic and Cl information can be used to increase compound identification confidence. False positives from an El‐only approaches can easily be identified and often correctly annotated using the additional information generated by Cl. Furthermore, compounds not listed at all in libraries have a high uncertainty for identification using an El‐only approach. Using accurate mass information on the molecular ion provided by Cl, sum formula for these unknowns can be derived. Combining this molecular information with the structural information generated by El, tentative structure elucidation becomes feasible in many cases.

Metabolite identification is essential in drug metabolismand pharmacokinetics(DMPK) studies and plays a pivotal role throughout the drug developmentprocess, from informing drug design to evaluating safety and efficacy.Mass spectrometry (MS) is the analytical technique of choice for characterizingmetabolites due to its selectivity and sensitivity, particularly whenpaired with chromatographic methods. However, MS encounters challengesin structural characterization. This study employs infrared ion spectroscopy(IRIS) to differentiate isomeric compounds and demonstrates the robustnessof a newly developed IRIS platform. We showcase applications in metaboliteidentification by determining the site of glucuronidation and phaseI oxidation in selected drug molecules. Employing density functionaltheory for spectral prediction, IRIS decreases reliance on referencestandards and alleviates the time-consuming purification processestypically associated with metabolite analysis. The newly developedplatform integrates a high-power, high-repetition-rate infrared laserand ion trap MS. This setup is very robust, as evidenced by the highlyreproducible IRIS spectra recorded over a one-year period withoutany instrument readjustment or recalibration. Moreover, the high powerand high repetition rate of the laser provide a large dynamic rangethat is necessary to resolve all spectral features. These resultsleverage IRIS toward a transformative tool in analytical chemistry,with potential applications expanding across various fields, suchas impurity analysis and forensics. The introduction of a compactIRIS setup in an industrial setting not only confirms its practicalapplicability but also emphasizes its potential for integration intoroutine analytical workflows.

Candida (Candidozyma) auris has distinct genetic clades. Clade distinction is relevant for infection control and epidemiological purposes. State-of-the-art typing methodologies for clade distinction are based on genomic approaches, such as short tandem repeat (STR) analysis and whole-genome sequencing (WGS). However, they are time-consuming and expensive. Fourier transform infrared spectroscopy (FT-IR) is an alternative tool for strain typing based on their unique biochemical spectral profiles. To apply FT-IR to differentiate C. auris clades and evaluate epidemiological relationships based on biochemical data among isolates from a multicenter C. auris outbreak in the state of Pernambuco, northeastern Brazil. Sixty-nine C. auris strains from clades I, II, III, and IV were analysed. Fifty-nine were clade IV strains obtained from three outbreaks that took place in Pernambuco state, northeastern Brazil. An adjusted FT-IR spectroscopy protocol was applied to obtain carbohydrates and lipid fingerprints. Short Tandem Repeat (STR) analysis was used in order to validate the spectroscopy approach. The adjusted preparation protocol for FT-IR analysis improved the spectral quality by 31.42% compared to standard protocol. FT-IR allowed us to discriminate C. auris clades I to IV. Moreover, important similarities were observed in C. auris clade IV strains obtained from two of the three hospitals, based on polysaccharides (1300-800 cm-1) plus lipids (3000-2800 cm-1 and 1500-1400 cm-1) spectra. STR confirmed the similarity results obtained by FT-IR, clustering the strains from two different hospitals. The IR Biotyper is fast, easy-to-use, and a promising alternative for moderate-to-high-complexity laboratories to differentiate C. auris clades. Furthermore, this technique has the potential for isolate-level source tracking, which could be valuable for monitoring transmission routes in clinical settings.

Carbon storage in soils is important in regulating atmospheric carbon dioxide (CO2). However, the sensitivity of the soil-carbon turnover time (τsoil) to temperature and hydrology forcing is not fully understood. Here, we use radiocarbon dating of plant-derived lipids in conjunction with reconstructions of temperature and rainfall from an eastern Mediterranean sediment core receiving terrigenous material from the Nile River watershed to investigate τsoilin subtropical and tropical areas during the last 18,000 years. We find that τsoil was reduced by an order of magnitude over the last deglaciation and that temperature was the major driver of these changes while the impact of hydroclimate was relatively small. We conclude that increased CO2 efflux from soils into the atmosphere constituted a positive feedback to global warming. However, simulated glacial-to-interglacial changes in a dynamic global vegetation model underestimate our data-based reconstructions of soil-carbon turnover times suggesting that this climate feedback is underestimated.

Abstract The rapidly evolving field of spatial biology revolves around the analysis of cells in their native microenvironment. This analysis can include morphological features, the presence of specific antigens or gene expression. To add another layer of information, recent methodological advances in matrix-assisted laser desorption ionization mass spectrometry imaging (MALDI-MSI) now enable the untargeted analysis of lipids and metabolites at subcellular resolution. The integration of MALDI-MSI at the single-cell level with established optical modalities, however, relies on an accurate yet intricate co-registration. Here, we describe the integration of bright-field and fluorescence microscopy into a prototype ion source of a state-of-the art MALDI-MSI instrument to obtain lipid and fluorescence microscopy-derived information from the same specimen, hence with intrinsic spatial correlation. We demonstrate the potential of the combined mass spectrometric and optical single-cell analysis on three examples. This includes the visualization of intracellular lipid distributions in macrophages, the introduction of pre-MALDI immunofluorescence staining on the example of murine cerebellum, and the heterogeneity of lipid profiles of tumor infiltrating neutrophils correlated to their individual microenvironments. Overall, the achieved tight correlation of single-cell lipid profiles with morphologic features and protein expression patterns constitutes a powerful resource for cell biology.

ZusammenfassungDer Klimawandel wird als eine der größten Herausforderungen der Gesellschaft in den nächsten Jahrzehnten bis Jahrhunderten gesehen. Ein Fokus gesetzgeberischer Aktivitäten liegt daher auf dem Themenkomplex der Nachhaltigkeit. So wurde auf EU-Ebene im Januar 2023 eine umfassende nichtfinanzielle Berichtspflicht etabliert. Im Bereich der Nachhaltigkeit in Lieferketten sind sowohl die EU als auch der nationale Gesetzgeber aktiv. Hieraus ergeben sich bereits jetzt und in Zukunft erst recht auch Rechtspflichten für die Akteure des Gesundheitswesens. Zudem zeichnet sich aufgrund einer EU-Richtlinie ein Kritis-Dachgesetz ab, das Akteure des Gesundheitswesens zu Schutzmaßnahmen vor Naturereignissen verpflichten wird. Hierüber vermittelt der Beitrag einen Überblick. Er geht außerdem auf eine mögliche Erweiterung des SGB V im Bereich der Nachhaltigkeit ein.