Biology subjects

Popova, E.

Publications and source records attributed to Popova, E..

5 recordsLinked to original sources

bioRxiv2025

Leveraging Multimodal Large Language Models to Extract Mechanistic Insights from Biomedical Visuals: A Case Study on COVID-19 and Neurodegenerative Diseases

BackgroundThe COVID-19 pandemic has intensified concerns about its long-term neurological impact, with growing evidence linking SARS-CoV-2 infection to neurodegenerative diseases (NDDs) such as Alzheimers (AD) and Parkinsons (PD). Patients with these conditions not only face higher risk of severe COVID-19 outcomes but may also undergo accelerated cognitive and motor decline following infection. Proposed mechanisms--ranging from neuroinflammation and blood-brain barrier disruption to abnormal protein aggregation--closely mirror core features of neurodegenerative pathology. Yet, current knowledge is fragmented across text, figures, and pathway diagrams, hindering integration into computational models capable of uncovering systemic patterns. ResultsTo address this gap, we applied GPT-4 Omni (GPT-4o), a multimodal large language model, to extract mechanistic insights from biomedical figures. Over 10,000 images were retrieved through targeted searches on COVID-19 and neurodegeneration; after automated and manual filtering, a curated subset was analyzed. GPT-4o extracted biological relationships as semantic triples, which were grouped into six mechanistic categories--including microglial activation and barrier disruption--using ontology-guided similarity and assembled into a Neo4j knowledge graph. Accuracy was evaluated against a gold-standard dataset of expert-annotated images using BioBERT-based semantic matching. This evaluation also enabled prompt tuning, threshold optimization, and hyperparameter assessment. Results demonstrate that GPT-4o successfully recovers both established and novel mechanisms, yielding interpretable outputs that illuminate complex biological links between SARS-CoV-2 and neurodegeneration. ConclusionsThis study showcases the potential of multimodal LLMs to mine biomedical visual data at scale. By complementing text mining and integrating figure-derived knowledge, our framework advances understanding of COVID-19-related neurodegeneration and supports future translational research.

bioinformatics↗

bioRxiv2025

Epigenetic modifiers to treat retinal degenerative diseases

We have previously demonstrated the ability of inhibitors of LSD1 and HDAC1 to block rod degeneration, preserve vision, maintain rod-specific transcripts and downregulate those involved in inflammation, gliosis, and cell death in the rd10 mouse model of Retinitis Pigmentosa (RP). To extend our findings we tested the hypothesis that this effect was due to altered chromatin structure by using a range of inhibitors of chromatin condensation to prevent photoreceptor degeneration in the rd10 mouse model. We used inhibitors for G9A/GLP that catalyzes methylation of H3K9, for EZH2 that catalyzes trimethylation of H3K27, and compared them to the actions of inhibitors of LSD1 and HDAC. All the inhibitors decondense chromatin and all preserve, to different extents, retinas from degeneration in rd10 mice, but they act through different metabolic pathways. One group of inhibitors, modifiers for LSD1 and EZH2, demonstrate a high level of maintenance of rod-specific transcripts, activation of Ca+2 and Wnt signaling pathways with inhibition of antigen processing and presentation, immune response and microglia phagocytosis. Another group of inhibitors, modifiers for HDAC and G9A/GLP work through upregulation of NGF-stimulated transcription, while down-regulating genes belong to immune response, extracellular matrix, cholesterol signaling and programmed cell death. Our results provide robust support for our hypothesis that inhibition of chromatin condensation can be sufficient to prevent rod death in rd10 mice.

neuroscience↗

bioRxiv2025

Spatialproteomics - an interoperable toolbox for analyzing highly multiplexed fluorescence image data

SummaryHighly multiplexed immunofluorescence imaging is a recent method to characterize tissues at single-cell resolution on the protein level, offering low cost, high scalability, and the ability to analyze paraffin-embedded tissue samples. However, the analysis of these data involves a sequence of steps, including segmentation, image processing, marker quantification, cell type classification, and neighborhood analysis, each of which involves a multitude of method and parameter choices that need to be adapted to the data and analytical objective at hand. Moreover, variations in data quality can be high and unpredictable, which necessitates further flexibility and interactivity. While individual components exist, there is an unmet need for a coherent toolbox that offers end-to-end coverage of the workflow, flexibility, and automation. We present spatialproteomics, a Python package that addresses these challenges. Built on top of xarray and dask, spatialproteomics can process images that are larger than the working memory. It supports synchronization of shared coordinates across data modalities such as images, segmentation masks, and expression matrices, which facilitates easy and safe subsetting and transformation. We demonstrate spatialproteomics on a set of images of reactive lymph nodes or different forms of B cell Non-Hodgkin lymphomas (BNHL) from 132 patients. We showcase an end-to-end analysis from raw images to statistical characterization of cell type composition and spatial distribution across indolent and aggressive lymphomas. Furthermore, we show how spatialproteomics can process gigapixel whole slide images. Altogether, we propose spatialproteomics as an easy-to-install, easy-to-learn, comprehensive toolbox for constructing powerful end-to-end image analysis solutions for highly multiplexed immunofluorescence imaging. Availability and ImplementationThe source code for spatialproteomics is freely available at https://github.com/sagar87/spatialproteomics under the MIT license. Contactwolfgang.huber@embl.org, Peter-Martin.Bruch@med.uni-duesseldorf.de

bioinformatics↗

bioRxiv2024

ATF4 orchestrates IL-1α-induced senescence in adult neural stem cells

Adult neural stem cells (NSC) are a potential source for the regeneration of damaged tissue during neuropathological conditions, but much remains unexplored. In an attempt to study the influence of neuroinflammation on NSCs, we generated a transgenic reporter rat strain that expresses the Discosoma sp. red (DsRed) fluorophore in NSCs and subjected it to traumatic brain injury (TBI). Transcriptomic analysis of NSCs isolated from TBI revealed an enrichment of stress response genes that pertained to endoplasmic reticulum (ER) stress and integrated stress response (ISR). Downstream analysis on NSC cultures pinpointed IL-1 as a trigger of ISR in these cells. At concentration levels similar to the ones measured post-TBI in rats, IL-1 induced the translation of activating transcription factor 4 (ATF4), an ISR master regulator. Further, ATF4 was necessary for the IL-1 -dependent induction of a senescent profile in NSCs, which included a metabolic shift towards glycolysis, induction of senescence-associated secretory phenotype, SASP, and cell cycle arrest. In summary, the ISR/ATF4 pathway seems to play a major role in NSC function during neuroinflammation and provides a therapeutic tool for protecting the NSC pool during these conditions.

neuroscience↗

bioRxiv2023

Karyopherin α2 is a maternal effect gene required for early embryonic development and female fertility in mice

The nuclear transport of proteins plays an important role in mediating the transition from egg to embryo and distinct karyopherins have been implicated in this process. Here, we studied the impact of KPNA2 deficiency on preimplantation embryo development in mice. Loss of KPNA2 results in complete arrest at the 2cell stage and embryos exhibit the inability to activate their embryonic genome as well as a severely disturbed nuclear translocation of Nucleoplasmin 2. Our findings define KPNA2 as a new maternal effect gene.

developmental biology↗