Unpacking Unstructured Data: A Pilot Study on Extracting Insights from Neuropathological Reports of Parkinson's Disease Patients using Large Language Models

Stroganov, O.; Schedlbauer, A.; Lorenzen, E.; Kadhim, A.; Lobanova, A.; Lewis, D. A.; Glausier, J. R.

doi:10.1101/2023.09.12.557252

Home/Archive/bioRxiv

bioRxiv · 10.1101/2023.09.12.557252

Unpacking Unstructured Data: A Pilot Study on Extracting Insights from Neuropathological Reports of Parkinson's Disease Patients using Large Language Models

Stroganov, O.·Schedlbauer, A.·Lorenzen, E.·Kadhim, A.·Lobanova, A.·Lewis, D. A.·Glausier, J. R.

View original source ↗Export metadata ↓

Abstract

ObjectiveThe aim of this study was to make unstructured neuropathological data, located in the NeuroBioBank (NBB), follow FAIR principles, and investigate the potential of Large Language Models (LLMs) in wrangling unstructured neuropathological reports. By making the currently inconsistent and disparate data findable, our overarching goal was to enhance research output and speed. Materials and MethodsThe NBB catalog currently includes information from medical records, interview results, and neuropathological reports. These reports contain crucial information necessary for conducting in-depth analysis of NBB data but have multiple formats that vary across different NBB biorepositories and change over time. In this study we focused on a subset of 822 donors with Parkinsons disease (PD) from 7 NBB biorepositories. We developed a data model with combined Brain Region and Pathological Findings data at its core. This approach made it easier to build an extraction pipeline and was flexible enough to convert resulting data to Common Data Elements (CDEs), a standardized data collection tools used by the neuroscience community to improve consistency and facilitate data sharing across studies. ResultsThis pilot study demonstrated the potential of LLMs in structuring unstructured neuropathological reports of PD patients available in the NBB. The pipeline enabled successful extraction of detailed tissue-level (microscopic) and gross anatomical (macroscopic) observations, along with staging information from pathology reports, with extraction quality comparable to manual curation results. To our knowledge, this is the first attempt to automatically standardize neuropathological information at this scale. The collected data has the potential to serve as a valuable resource for PD researchers, facilitating integration with clinical information and genetic data (such as genome-wide genotyping and whole genome sequencing) available through the NBB, hereby enabling a more comprehensive understanding of the disease.

Source connections

Published article ↗

Explore related subjects

pathology Preprint

Keep this discovery

Explore connections, maps & timelines

BibTeX RIS

Stroganov, O., Schedlbauer, A., Lorenzen, E., Kadhim, A., Lobanova, A., Lewis, D. A., Glausier, J. R.. 2023-09-15. Unpacking Unstructured Data: A Pilot Study on Extracting Insights from Neuropathological Reports of Parkinson's Disease Patients using Large Language Models. https://doi.org/10.1101/2023.09.12.557252

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related preprints

bioRxiv2026

NAE1-Dependent Protein Neddylation Preserves Endothelial Identity and Vascular Integrity

Background: Endothelial dysfunction is a central driver of cardiovascular and inflammatory diseases, yet the post-translational mechanisms that preserve endothelial homeostasis remain incompletely understood. Protein neddylation, the covalent conjugation of a ubiquitin-like modifier, regulates diverse cellular processes, yet its physiological role in the vascular endothelium remains unknown. This study investigated whether protein neddylation is required to preserve endothelial identity and vascular homeostasis. Methods: We generated tamoxifen-inducible endothelial-specific Nae1 knockout mice to inhibit neddylation and combined bulk RNA sequencing, single-cell and single-nucleus transcriptomics, quantitative proteomics, biochemical analyses, and gain- and loss-of-function approaches to define the role of endothelial neddylation in vascular homeostasis and inflammatory injury. Results: Endothelial-specific Nae1 deletion caused rapid mortality associated with vascular leakage, platelet accumulation, inflammation, and multi-organ injury. Multi-omics analyses demonstrated profound loss of endothelial identity, characterized by suppression of core endothelial programs and activation of inflammatory, procoagulant, and pyroptotic pathways. Single-cell analyses revealed progressive endothelial dysfunction culminating in depletion of the endothelial population and remodeling of the vascular niche. Mechanistically, endothelial neddylation deficiency activated gasdermin D (GSDMD)- and gasdermin E (GSDME)-dependent pyroptosis, whereas dual inhibition of GSDMD and GSDME markedly attenuated inflammatory transcriptomic remodeling, vascular injury, hepatocyte death, immune cell infiltration, and platelet accumulation. Translational analyses demonstrated reduced endothelial neddylation in experimental endotoxemia and decreased expression of neddylation pathway components in human atherosclerosis and COVID-19 datasets. Conversely, restoration of endothelial neddylation partially reversed inflammatory endothelial transcriptomic reprogramming in vivo. Conclusions: NAE1-dependent protein neddylation is an essential regulator of endothelial identity and vascular integrity. Loss of endothelial neddylation promotes gasdermin-dependent pyroptosis and thrombo-inflammatory vascular injury, whereas restoration of the neddylation pathway mitigates inflammatory endothelial dysfunction. These findings identify endothelial neddylation as a fundamental mechanism maintaining vascular homeostasis and a potential therapeutic target for cardiovascular and inflammatory diseases.

pathology↗

bioRxiv2025

Pregnant Dairy Heifers Express Influenza A Virus Receptors in the Mammary Gland

Highly pathogenic avian influenza (HPAI) A(H5N1) virus emerged in lactating dairy cattle in March 2024, causing mastitis-related disease and infections in other farm animals and workers. Recent work identified 2,6 and 2,3-linked sialic acids (SA), which serve as influenza virus receptors, in the lactating bovine mammary gland; however, their distribution across stages of mammary growth and development remains unknown. We compared the distribution of tissue sialylation in mammary glands of prepubertal dairy calves, pregnant dairy heifers, and lactating cows. Mammary glands at all physiological stages expressed both 2,6 SA, the preferred receptor linkage for human influenza viruses, and 2,3 SA, the preferred receptor linkage for avian influenza viruses. Importantly, mammary glands of pregnant dairy heifers exhibited the highest overall expression of 2,3 SA, observed in both tissue and alveolar lumens. Our results suggest that pregnant dairy heifers, like lactating dairy cows, are susceptible to H5N1 infection in the mammary gland.

pathology↗

bioRxiv2025

Latent transforming growth factor binding protein-2 (LTBP2), an IPF biomarker of clinical decline, promotes TGF-beta signaling and lung fibrosis in mice

The identification of clinically predictive serum biomarkers for pulmonary fibrosis is a significant challenge and important goal. Multiple recent proteomic biomarker studies have identified latent transforming growth factor binding protein-2 (LTBP2) as a circulating factor associated with disease progression in fibrotic lung diseases in humans (including IPF), but its role in the development of fibrosis is incompletely defined. LTBP2 competes with the large latent transforming growth factor-beta (TGF{beta}) complex (LLC) for binding to the N-terminus of fibrillin and is thought to promote the release of active TGF{beta}. We hypothesized that LTBP2 deficiency would promote LLC sequestration in matrix and reduce TGF{beta} signaling. We recently reported an LTBP2 knockout (Ltbp2-/-) mouse with no baseline lung abnormalities. Here we show that Ltbp2-/- mice exposed to either bleomycin or silica have a significant reduction in fibrosis compared to wild type controls. Consistent with reduced fibrosis, after bleomycin Ltbp2-/- mouse lungs have reduced TGF{beta} signaling and isolated fibroblasts from Ltbp2-/-mice exhibit impaired migration in an in vitro wound closure assay. Transcriptomic analysis of bleomycin-treated control and Ltbp2-/- mouse lung tissue identified multiple LTBP2-regulated genes, including the lncRNA antisense of IGFR2 non-coding RNA (Airn) which has reported antifibrotic effects. Interestingly, we also observed that Ltbp2-/- mice had impaired epithelial repair after bleomycin treatment, a phenotype that also occurred in a naphthalene model of club cell injury. These findings provide evidence that LTBP2 is profibrotic and facilitates TGF{beta} signaling but is also required for normal airway epithelial repair.

pathology↗