New Research Reveals Genetic Clues to Long COVID Subtypes

A new study in PLOS Computational Biology employs a large-scale “multi-omics” approach, integrating genetic data, gene-expression information, and network analysis to better understand why Long Covid affects people in so many different ways.

Using this integrative framework, the researchers identified 32 “putative causal genes” that may help drive Long Covid.

Nineteen of these genes had already been reported in prior COVID-19 research, lending confidence to the results. These genes point to several biological processes that may be involved, including immune regulation, viral carcinogenesis (how viruses can influence cell behavior), cell-cycle control (how cells grow, divide, and repair themselves), and metabolic adaptation (how cells adjust under stress).

The study also grouped patients by linking their symptom profiles to gene-expression clustering, revealing three “biologically meaningful subtypes” of Long COVID.

In simple terms, people tended to fall into symptom clusters that matched underlying gene-expression patterns, helping explain why Long Covid can look so different from one person to another.

The authors suggest that understanding Long Covid through these subtypes could help future research and clinical trials become more targeted, matching treatments to the biology driving each group rather than treating Long Covid as one single condition.

Citation

Pinero, S., Li, X., Liu, L., Li, J., Lee, S. H., Winter, M., Nguyen, T., Zhang, J., & Le, T. D. (2025). Integrative multi-omics framework for causal gene discovery in Long COVID. PLoS computational biology, 21(12), e1013725. https://doi.org/10.1371/journal.pcbi.1013725