Integrative machine learning identifies robust inflammation-related diagnostic biomarkers and stratifies immune-heterogeneous subtypes in Kawasaki disease
Authors' conclusion
This study establishes inflammation-related diagnostic biomarkers and immune-stratified subtypes for KD
Study context
- Population
- Kawasaki disease (KD), a pediatric systemic vasculitis; multi-cohort transcriptomic data
- Comparison
- Cluster1 vs Cluster2
- Outcome
- Inflammation-related diagnostic biomarkers of KD
Methodological detailsmulti-cohort transcriptomic integration, WGCNA, machine learning (LASSO, Boruta, SVM-RFE, Random Forest), consensus clustering
- Multi-cohort transcriptomic integration
- WGCNA
- Machine learning (LASSO, Boruta, SVM-RFE, Random Forest)
- Consensus clustering
Main finding
As reported in the abstract
Diagnostic biomarkers
Six robust diagnostic biomarkers (ADM, ALPL, FCGR1A, HP, S100A12, SLC22A4) identified; AUC in cohorts
> 0.9
Further findings
Immune infiltration in KD
- Elevated neutrophils, monocytes and Tregs; reduced CD8 + T cells and cytolytic activity
Molecular subtypes (consensus clustering)
- Two immune-heterogeneous subtypes: Cluster1 (neutrophil/Treg-dominant, TLR signaling) and Cluster2 (B cell/CD8 + T cell-dominant, cytolytic activity)
- Regulatory networks revealed subtype-specific transcriptional regulators and therapeutic agents
Original publication
Integrative machine learning identifies robust inflammation-related diagnostic biomarkers and stratifies immune-heterogeneous subtypes in Kawasaki disease. Pediatr Rheumatol Online J. Published 2 June 2025. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →