Gene set correlation enrichment analysis for interpreting and annotating gene expression profiles.

Area of Science:

• Genomics
• Bioinformatics
• Systems Biology

Background:

• Pathway analysis interprets gene expression differences between phenotypes.
• Topology-based (TB) methods often outperform non-TB methods by considering gene interactions.
• Existing TB methods face limitations with incomplete pathway data and reference network selection.

Purpose of the Study:

• To introduce Gscore, a novel gene set correlation enrichment analysis method.
• To evaluate Gscore's performance against state-of-the-art methods using diverse human disease expression datasets.
• To explore Gscore's utility in analyzing COVID-19 related gene expression data.

Main Methods:

• Developed Gscore based on an expression dataset-derived coexpression network.
• Assessed whether differentially expressed gene (DEG) lists or individual DEGs associate with known gene sets.
• Compared Gscore against eight other methods across 89 human disease expression datasets.

Main Results:

• Gscore demonstrated superior performance in identifying target pathways compared to eight other methods.
• The method provides insights into disease-wide and pathway-wide associations with clinical outcomes.
• Application to COVID-19 data revealed implications of DEGs in relevant pathways.

Conclusions:

• Gscore offers a powerful analytical approach for annotating DEGs and expression profiles.
• The method enhances biological interpretation by leveraging coexpression networks.
• Gscore facilitates the understanding of gene-disease associations and clinical relevance.