Unraveling the Co-Morbidity between COVID-19 and Neurodegenerative Diseases Through Multi-scale Graph Analysis: A Systematic Investigation of Biological Databases and Text Mining

The COVID-19 pandemic has produced an overwhelming volume of research, yet much of it remains focused on individual diseases, largely ignoring the complex relationships between comorbidities. Although extensive literature exists on both neurodegenerative diseases (NDDs), namely Alzheimer’s and Parkinson’s, and COVID-19, their intersection remains underexplored. Co-morbidity modeling is essential, as patients, particularly those hospitalized, often present with multiple conditions. This study addresses this gap by investigating the crosstalk between COVID-19 and NDDs using a combination of knowledge graphs built from curated biomedical datasets and text mining tools. We conducted comprehensive graph analyses, including path analysis, phenotype coverage assessment, and mapping of cellular and genetic factors, to examine how various Knowledge Graphs (KGs)—such as PrimeKG, DrugBank, OpenTargets, and those generated from natural language processing (NLP) methodologies—illuminate molecular and phenotypic relationships between these diseases. Our findings reveal significant variability in graph density and connectivity across datasets, each offering unique insights into the landscape of COVID-19 and NDD co-morbidities. By integrating structured biological data with unstructured textual data, this study aimed to optimize co-morbidity modeling, maximize recall for identifying potential co-morbidity mechanisms, and consolidate this information in a dedicated co-morbidity hypothesis database. Key genetic and inflammatory markers, particularly immune response genes, emerged as consistent features across multiple KGs, reinforcing their potential role in COVID-19–NDD interactions. This integrative approach advances our understanding of the underlying mechanisms linking these diseases and facilitates the identification of potential therapeutic targets.

All data, methodologies, and detailed instructions for accessing the co-morbidity hypothesis database are publicly available and thoroughly documented at: https://github.com/SCAI-BIO/covid-NDD-comorbidity-NLP .