Exploring hypoxia driven subtypes of pulmonary arterial hypertension through transcriptomics single cell sequencing...

Area of Science:

• Cardiovascular Research
• Genomics and Bioinformatics
• Machine Learning in Medicine

Background:

• Pulmonary arterial hypertension (PAH) is a severe cardiovascular disease with poor prognosis and unclear mechanisms.
• Current treatments for PAH offer limited efficacy, highlighting the need for improved diagnostic and therapeutic strategies.

Purpose of the Study:

• To integrate multi-omics data and machine learning to identify molecular subtypes of PAH.
• To uncover distinct pathogenic mechanisms and potential therapeutic targets for personalized PAH treatment.

Main Methods:

• Consensus clustering of multi-omics data to classify PAH patients.
• Differential gene expression and pathway enrichment analyses.
• Development and validation of machine learning models for subtype prediction.

Main Results:

• PAH patients were classified into two subgroups (C1 and C2).
• Subgroup C2 displayed significantly upregulated hypoxia-related genes, with enriched TNF, IL-17, and HIF-1 pathways.
• Machine learning models accurately distinguished hypoxia-associated subtypes (AUC > 0.85).

Conclusions:

• Novel molecular subtypes of PAH with distinct mechanisms, particularly involving hypoxia, were identified.
• Key genes and pathways offer potential diagnostic biomarkers and therapeutic targets for PAH.
• Machine learning aids in precise PAH diagnosis and subtype classification.