qusage

qusage performs quantitative set analysis for gene expression by estimating gene-set activity and accounting for inter-gene correlation to improve gene set enrichment analysis of genome-wide expression data.

Key Features:

• Inter-Gene Correlation Handling: Accounts for inter-gene correlations using a data-derived variance inflation factor (VIF) to address elevated Type I error rates in traditional methods like GSEA.
• Probability Density Function (PDF) Representation: Quantifies gene-set activity using a full probability density function rather than relying solely on P-values.
• Confidence Intervals and Post Hoc Analysis: Extracts P-values and confidence intervals from the PDF to enable post hoc comparisons with preserved statistical traceability.
• Comparative Performance: Demonstrated higher sensitivity and specificity than GSEA and CAMERA in studies profiling interferon therapy in chronic Hepatitis C virus patients and Influenza A virus infection.
• R Package Implementation: Implemented as an R package that provides core analysis functions and plotting/visualization routines.

Scientific Applications:

• Functional Genomics: Provides quantitative gene-set activity estimates for functional genomics analyses of genome-wide expression data.
• Clinical Transcriptomics: Applied to profiling therapeutic responses, exemplified by interferon therapy in chronic Hepatitis C virus patients.
• Infectious Disease Transcriptomics: Applied to host response analyses, exemplified by Influenza A virus infection studies.
• Hypothesis Testing and Discovery: Supports robust hypothesis testing and discovery in genomic studies through interval estimates and post hoc comparisons.

Methodology:

Estimates a data-derived variance inflation factor (VIF) to account for inter-gene correlation; represents gene-set activity as a probability density function (PDF) and derives P-values and confidence intervals from that PDF for post hoc comparisons; provided as an R package implementation.