Integrated Single-Cell Atlas Reveals Cross-Subtype Heterogeneity in Human Pulmonary Hypertension.

Area of Science:

• Pulmonary vascular disease research
• Single-cell genomics
• Bioinformatics

Background:

• Pulmonary hypertension (PH) involves complex pulmonary vascular remodeling.
• Existing single-cell studies in PH are fragmented across subtypes and analysis pipelines.
• This fragmentation limits cross-study comparisons and understanding of disease-relevant cell states in human PH.

Purpose of the Study:

• To create a unified, high-quality single-cell RNA sequencing atlas of human PH.
• To enable standardized cross-subtype comparisons of cellular alterations in PH.
• To develop a computational tool for efficient cell type annotation in PH studies.

Main Methods:

• Integrated multiple single-cell RNA sequencing datasets using standardized pipelines.
• Constructed a core PH-specific cellular atlas with hierarchical cell type annotation.
• Developed and validated PH-Map, a multitask learning tool for cell annotation.

Main Results:

• Generated a core atlas of 235,621 cells from 64 human PH samples across subtypes and controls.
• Identified shared and subtype-specific cellular remodeling in vascular and immune compartments.
• PH-Map demonstrated stable and efficient hierarchical annotation performance in external cohorts.

Conclusions:

• The human single-cell PH atlas facilitates systematic cross-subtype analysis.
• The atlas delineates shared and specific cellular programs in PH vascular and immune cells.
• The atlas and PH-Map tool support reproducible annotation and pathway prioritization for PH research.