Coherent pathway enrichment estimation by modeling inter-pathway dependencies using regularized regression

Area of Science:

• Bioinformatics
• Computational Biology
• Genomics

Background:

• Gene set enrichment analysis (GSEA) interprets gene lists from differential expression studies.
• Existing GSEA tools use pathway databases (KEGG, Reactome, GO).
• Database size and pathway redundancy complicate GSEA.

Purpose of the Study:

• To develop a novel GSEA method addressing pathway dependencies.
• To improve robustness and interpretability of gene set enrichment analysis.

Main Methods:

• Developed pareg, a GSEA method using a regularized generalized linear model.
• Incorporated dependencies between gene sets in the enrichment computation.
• Utilized breast cancer samples from The Cancer Genome Atlas (TCGA) for exploratory analysis.

Main Results:

• pareg demonstrates increased robustness to noise compared to existing methods.
• The method successfully recovered known biological pathways.
• pareg identified potential novel treatment targets in breast cancer.

Conclusions:

• pareg offers a more robust and accurate approach to gene set enrichment analysis.
• The method effectively handles complex pathway relationships.
• pareg has potential applications in identifying therapeutic strategies.