Unique and shared transcriptomic signatures underlying localized scleroderma pathogenesis identified using interpretable machine learning DOI Creative Commons
Aaron B.I. Rosen, Anwesha Sanyal,

Theresa Hutchins

et al.

JCI Insight, Journal Year: 2025, Volume and Issue: 10(7)

Published: April 7, 2025

Using transcriptomic profiling at single-cell resolution, we investigated cell-intrinsic and cell-extrinsic signatures associated with pathogenesis inflammation-driven fibrosis in both adult pediatric patients localized scleroderma (LS). We performed RNA-Seq on LS healthy controls. then analyzed the data using an interpretable factor analysis machine learning framework, significant latent interaction discovery exploration (SLIDE), which moves beyond predictive biomarkers to infer factors underlying pathophysiology. SLIDE is a recently developed regression-based framework that comes rigorous statistical guarantees regarding identifiability of factors, corresponding inference, FDR control. found distinct differences characteristics complexity molecular between LS. identified cell type-specific determinants age severity revealed insights into signaling mechanisms shared systemic sclerosis (SSc), as well onset disease compared population. Our analyses recapitulate known drivers pathology identify cellular modules stratify subtypes define axis SSc.

Language: Английский

Unique and shared transcriptomic signatures underlying localized scleroderma pathogenesis identified using interpretable machine learning DOI Creative Commons
Aaron B.I. Rosen, Anwesha Sanyal,

Theresa Hutchins

et al.

JCI Insight, Journal Year: 2025, Volume and Issue: 10(7)

Published: April 7, 2025

Using transcriptomic profiling at single-cell resolution, we investigated cell-intrinsic and cell-extrinsic signatures associated with pathogenesis inflammation-driven fibrosis in both adult pediatric patients localized scleroderma (LS). We performed RNA-Seq on LS healthy controls. then analyzed the data using an interpretable factor analysis machine learning framework, significant latent interaction discovery exploration (SLIDE), which moves beyond predictive biomarkers to infer factors underlying pathophysiology. SLIDE is a recently developed regression-based framework that comes rigorous statistical guarantees regarding identifiability of factors, corresponding inference, FDR control. found distinct differences characteristics complexity molecular between LS. identified cell type-specific determinants age severity revealed insights into signaling mechanisms shared systemic sclerosis (SSc), as well onset disease compared population. Our analyses recapitulate known drivers pathology identify cellular modules stratify subtypes define axis SSc.

Language: Английский

Citations

0