Biology subjects

Mosig, A.

Publications and source records attributed to Mosig, A..

6 recordsLinked to original sources

bioRxiv2026

ToxCastLite: A portable semantic evidence graph linking in vitro bioactivity, in vivo toxicity, and exposure-use context

MotivationThe ToxCast database is a valuable resource for computational toxicology and new approach methodologies (NAMs), but the approximately 100 GB MySQL distribution is difficult to use for portable local analysis and cross-domain evidence mining. Many practical questions concern chemicals, in vitro bioactivity, in vivo toxicological evidence, and exposure-relevant product-use context rather than raw database keys. ResultsWe present ToxCastLite, a portable semantic evidence-access system that combines assay-scoped SQLite databases with a compact RDF layer for GraphDB-based querying. The system streams large ToxCast/invitrodb MySQL dumps into curated SQLite profiles, reducing the footprint to approximately 3 GB for focused use cases such as developmental neurotoxicity. Dense numerical evidence, including concentration-response rows, remains in SQLite, while the RDF projection exposes linked semantic entities such as chemicals, assays, endpoints, model results, potency parameters (AC50), and MC6 quality flags. We further extend the graph with CPDat v4.0 product-use and functional-use evidence and ToxRefDB v3.0 in vivo toxicity evidence, including processed studies, point-of-departure records, effect summaries, and observation summaries. These layers are linked through DSSTox Substance Identifiers, enabling integrated queries across NAM bioactivity, curated animal-study evidence, and exposure/use context. A Streamlit prototype supports exploration through a locally deployed LLM that translates natural-language questions into SPARQL, grounded by a versioned RDF schema to reduce hallucination risk. Case studies in developmental neurotoxicity demonstrate how ToxCastLite identifies concordance between high-confidence in vitro DNT activity and positive in vivo apical evidence, detects in vitro DNT activity beyond available DNT-specific in vivo evidence, and prioritizes chemicals where NAM signals, ToxRefDB evidence, and CPDat product-use context intersect. For selected results, users can drill down from the semantic graph to the underlying SQLite records and retrieve concentration-response curves for expert inspection without manually writing SQL or SPARQL. AvailabilityProject website at toxcast-lite.github.io/. Contactarif.doenmez@iuf-duesseldorf.de

bioinformatics↗

bioRxiv2022

Biostatistics and its impact on hazard characterization using in vitro developmental neurotoxicity assays

In the field of hazard assessment, Benchmark concentrations (BMC) and their associated uncertainty are of particular interest for regulatory decision making. The BMC estimation consists of various statistical decisions to be made, which depend largely on factors such as experimental design and assay endpoint features. In current data practice, the experimenter is often responsible for the data analysis and therefore relies on statistical software without being aware about the software default settings and how they can impact the outputs of data analysis. To provide more insight into how statistical decision making can influence the outcomes of data analysis and interpretation, we have used case studies on a large dataset produced by a developmental neurotoxicity (DNT) in vitro battery (DNT IVB). Here we focused on the BMC and its confidence interval (CI) estimation, as well as on the final hazard classification. We identified five crucial statistical decisions experimenter have to face during data analysis: choice of replicate averaging, response data normalization, regression modelling, BMC and CI estimation, as well as choice of benchmark response levels. In addition, the strength of our data evaluation platform is the integration of endpoint-specific hazard classifications, including flagging systems for uncertain cases, which none of the so far existing statistical data analysis platforms provide. The insights gained in this study demonstrate how important fit-for-purpose, internationally harmonized and accepted data evaluation and analysis procedures are for an objective hazard classification.

bioinformatics↗

bioRxiv2022

Rapid and fully automated blood vasculature analysis in 3D light-sheet image volumes of different organs

Blood vasculature represents a complex network of vessels with varying lengths and diameters that are precisely organized in space to allow proper tissue function. Light-sheet fluorescence microscopy (LSFM) is very useful to generate tomograms of tissue vasculature with high spatial accuracy. Yet, quantitative LSFM analysis is still cumbersome and available methods are restricted to single organs and advanced computing hardware. Here, we introduce VesselExpress, an automated software that reliably analyzes six characteristic vascular network parameters including vessel diameter in LSFM data on average computing hardware. VesselExpress is ~100 times faster than other existing vessel analysis tools, requires no user interaction, integrates batch processing, and parallelization. Employing an innovative dual Frangi filter approach we show that obesity induces a large-scale modulation of brain vasculature in mice and that seven other major organs differ strongly in their 3D vascular makeup. Hence, VesselExpress transforms LSFM from an observational to an analytical working tool.

bioinformatics↗

bioRxiv2021

Human macrophage polarization determines bacterial persistence of Staphylococcus aureus in a liver-on-chip-based infection model

Infections with Staphylococcus aureus (S. aureus) have been reported from various organs ranging from asymptomatic colonization to severe infections and sepsis associated with multiple organ dysfunction. Although considered an extracellular pathogen, S. aureus can invade and persist in professional phagocytes such as monocytes and macrophages. Its capability to persist and manipulate phagocytes is considered a critical step to evade host antimicrobial reactions. For the first time we leveraged a human liver-on-chip model and tailored image analysis algorithms to demonstrate that S. aureus (USA300) specifically targets macrophages in the liver models as essential niche facilitating bacterial persistence and phenotype switching to small colony variants (SCVs). In vitro M2 polarization was found to favor SCV-formation and was associated with increased intracellular bacterial loads in macrophages, increased cell death, and impaired recruitment of circulating monocytes to sites of infection. These findings expand the knowledge about the role of liver macrophages in the course of systemic infection. Further, the results might be relevant for understanding infection mechanisms in patients with chronic liver disease such as fibrosis that display increased frequencies of M2 polarized liver macrophages and have a higher risk for developing chronic infections and relapsing bacteremia.

microbiology↗

bioRxiv2020

UC2 - A Versatile and Customizable low-cost 3D-printed Optical Open-Standard for microscopic imaging

With UC2 (You-See-Too) we present an inexpensive 3D-printed microscopy toolbox. The system is based on concepts of modular development, rapid-prototyping and all-time accessibility using widely available off-the-shelf optic and electronic components. We aim to democratize microscopy, reduce the reproduction crisis and enhance trust into science by making it available to everyone via an open-access public repository. Due to its versatility the aim is to boost creativity and non-conventional approaches. In this paper, we demonstrate a development cycle from basic blocks to different microscopic techniques. First, we build a bright-field system and stress-test it by observing macrophage cell differentiation, apoptosis and proliferation incubator-enclosed for seven days with automatic focussing to minimize axial drift. We prove versatility by assembling a system using the same components to a fully working fluorescence light-sheet system and acquire a 3D volume of a GFP-expressing living drosophila larvae. Finally, we sketch and demonstrate further possible setups to draw a picture on how the system can be used for reproducible prototyping in scientific research. All design files for replicating the experimental setups are provided via an open-access online repository (https://github.com/bionanoimaging/UC2-GIT) to foster widespread use.

bioengineering↗

bioRxiv2019

Deep representation learning for domain adaptable classification of infrared spectral imaging data

MotivationApplying infrared microscopy in the context of tissue diagnostics heavily relies on computationally preprocessing the infrared pixel spectra that constitute an infrared microscopic image. Existing approaches involve physical models, which are non-linear in nature and lead to classifiers that do not generalize well, e.g. across different types of tissue preparation. Furthermore, existing preprocessing approaches involve iterative procedures that are computationally demanding, so that computation time required for preprocessing does not keep pace with recent progress in infrared microscopes which can capture whole-slide images within minutes.\n\nResultsWe investigate the application of stacked contractive autoencoders as an unsupervised approach to preprocess infrared microscopic pixel spectra, followed by supervised fine-tuning to obtain neural networks that can reliably resolve tissue structure. To validate the robustness of the resulting classifier, we demonstrate that a network trained on embedded tissue can be transferred to classify fresh frozen tissue. The features obtained from unsupervised pretraining thus generalize across the large spectral differences between embedded and fresh frozen tissue, where under previous approaches seperate classifiers had to be trained from scratch.\n\nAvailabilityOur implementation can be downloaded from https://github.com/arnrau/SCAE_IR_Spectral_Imaging\n\nContactaxel.mosig@bph.rub.de

bioinformatics↗