Back

Neuroimmune pleiotropy links COVID-19 outcomes to brain structural and functional imaging-derived phenotypes

Chen, Q.; He, J.; Cabrera Mendoza, B.; Xu, Z.; Qiu, D.; Polimanti, R.

2025-11-04 genetic and genomic medicine
10.1101/2025.11.03.25339411 medRxiv
Show abstract

This study investigated the pleiotropy and the underlying biology linking COVID-19 outcomes (infection, hospitalization, and severe symptoms) to brain structure and function, leveraging genome-wide data from the COVID-19 Host Genetics Initiative (up to 122,616 cases and 2,475,240 controls) and 3,935 brain imaging-derived phenotypes (IDP) from the UK Biobank (up to 33,224 participants). COVID-19 outcomes were genetically correlated with 11 IDPs related to brain connectivity, cortical structures, and white matter. With respect to these, we also identified local genetic correlations within five loci mapping genes involved in inflammatory response, immune regulation, brain development, and neuropsychiatric disorders. We also identified 305 Bonferroni-significant gene ontologies (GO), highlighting the pleiotropy linking COVID-19 to brain structure and function through developmental processes (e.g., immune system development, leukocyte differentiation, and neurogenesis) and intracellular signaling (e.g., synaptic signaling and immune response regulation). Our drug-repurposing analysis uncovered 41 compounds, including approved drugs that could potentially influence both COVID-19 outcomes and brain-related disorders. For instance, chlorpromazine is an antipsychotic that appears to also have antiviral and immunomodulatory activity. Overall, the present findings contributed to dissecting the biological mechanisms shared between COVID-19 and brain structure and function, highlighting the systemic nature of their relationship.

Matching journals

The top 2 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.