Related Experiment Video
Updated: Jun 25, 2026
14:58
High-Throughput Transcriptome Analysis for Investigating Host-Pathogen Interactions
Published on: March 5, 2022
Integrated miRNAome-transcriptome analyses identify an immuno-hematopoietic subcluster in patients with long COVID
Ying Yang1, Mari Kanerva2, Helena Liira3
1Institute of Environmental Medicine, Karolinska Institutet, Stockholm, Sweden; Human Microbiome Research Program, Faculty of Medicine, University of Helsinki, Helsinki, Finland.
The Journal of Allergy and Clinical Immunology
|June 23, 2026
Summary
Persistent symptoms after SARS-CoV-2 infection, known as long COVID (LC), show distinct molecular subtypes. Integrative profiling revealed an immune-hematopoietic LC subcluster (LC1) with significant clinical impairment, suggesting potential for biomarker-based patient stratification.
Area of Science:
- Molecular biology
- Immunology
- Genomics
Background:
- Persistent symptoms following SARS-CoV-2 infection, termed long COVID (LC), present significant burdens.
- The underlying molecular mechanisms, particularly miRNA-mRNA interactions, contributing to LC heterogeneity are not fully understood.
Purpose of the Study:
- To investigate if integrated blood miRNA and mRNA profiling can identify distinct molecular subclusters within long COVID.
- To link these molecular subclusters to specific clinical and immune-hematopoietic features.
Main Methods:
- Performed integrated miRNAome-transcriptome profiling of blood RNA from individuals with LC and controls.
- Utilized hierarchical clustering to define LC subclusters and constructed miRNA-mRNA interaction networks.
- Compared clinical, functional, and biochemical parameters between subclusters and developed a random forest classifier.
Main Results:
- Identified an immune-hematopoietic LC subcluster (LC1) with significant miRNA-mRNA dysregulation and pathway enrichment.
- LC1 exhibited greater symptom burden, functional impairment, reduced quality of life, and distinct biochemical alterations (e.g., lower sodium, elevated fibrin D-dimer).
- A random forest classifier achieved high accuracy (AUROC 0.91) in distinguishing LC1 from other groups.
Conclusions:
- Long COVID consists of biologically and clinically distinct subclusters driven by coordinated miRNA-mRNA remodeling.
- The identified immune-hematopoietic LC1 subtype is suitable for biomarker-based stratification of patients with persistent impairment.
