Dissecting complex transcriptional responses using pathway-level scores based on prior information

Area of Science:

• Genomics
• Systems Biology
• Bioinformatics

Background:

• Cellular responses to environmental changes create complex mRNA expression patterns.
• Integrating prior gene function or regulatory interaction data aids in dissecting transcriptional responses.

Purpose of the Study:

• To review and detail two methods for robust pathway-level differential gene expression analysis.
• To enhance the interpretation of complex genomewide mRNA expression data.

Main Methods:

• Comparing gene set expression distributions against the genome.
• Utilizing regression analysis on estimated gene regulatory network connectivities.
• Detailed explanation of statistical methodologies.

Main Results:

• Two distinct approaches for combining individual gene expression levels into pathway-level scores are presented.
• Methods allow for robust estimation of gene regulatory pathway activity.

Conclusions:

• Pathway-level analysis using prior information is more sensitive to subtle gene expression changes than gene-level analysis.
• The presented methods are statistically sound, interpretable, and technically straightforward.