Multi-omics and Machine Learning Exploration of Key Genes in Abdominal Aortic Aneurysm

Integrated multi-omics analysis from 2026 identifies candidate biomarkers for abdominal aortic aneurysm (AAA). The dataset contains results from merging two microarray datasets (GSE47472, GSE57691), validated with GSE7084, and applying Mendelian randomization and machine learning. It includes 551 differentially expressed genes, with four genes—PLAU, CD58, PCYOX1, and THBS4—prioritized as potential biomarkers.

Use Cases

• Biomarker validation based on Mendelian randomization and colocalization results
• Gene expression analysis using the list of 551 differentially expressed genes
• Machine learning model training for AAA risk prediction using the prioritized gene set
• Pathway enrichment analysis informed by immune-infiltration profiling results
• Cross-referencing candidate genes (e.g., PLAU) with external proteomic or genetic datasets

Strengths

• Dataset integrates genetic, transcriptomic, and proteomic evidence via a multi-omics framework
• Analysis validated in an independent cohort (GSE7084) and a murine AAA model
• Results include specific performance metrics (e.g., PLAU AUC of 0.854 and 0.944)

Limitations

• Column names and row counts are not provided, limiting understanding of data structure
• File sizes conflict slightly between sources (10002 vs 9723), indicating potential versioning issues

Provenance

Source

Integrated analysis of public microarray datasets (GSE47472, GSE57691, GSE7084).

Collection Method

Microarray data merged with batch correction, followed by Mendelian randomization, machine learning, and experimental validation.

Freshness

Last updated 2026-05-29