A novel method to prioritize RNAseq data for post-hoc analysis based on absolute changes in transcript abundance.

Area of Science:

• Transcriptomics
• Bioinformatics
• Systems Biology

Background:

• Fold-change (FC) is commonly used to identify differentially expressed genes (DEGs) in RNA sequencing (RNAseq) data.
• FC-based prioritization can miss biologically important DEGs, particularly high-copy-number transcripts with significant expression changes that don't meet ratiometric cut-offs.

Purpose of the Study:

• To evaluate an alternative method for prioritizing RNAseq data using absolute changes in normalized transcript counts (ΔT).
• To determine if prioritizing DEGs by ΔT magnitude identifies biologically meaningful transcriptional programs.

Main Methods:

• Analyzed five pairwise RNAseq comparisons with varying effect sizes.
• Rank-ordered DEGs based on the magnitude of absolute transcript count change (ΔT).
• Evaluated large ΔT gene sets for Gene Ontology (GO) and protein interaction enrichment.

Main Results:

• A small percentage of transcripts (4.7-5.0%) accounted for a substantial portion (36-50%) of cumulative expression change, characterized by large ΔT values.
• Large ΔT genes were significantly enriched for GO terms and protein interactions, consistent with biological context and distinct from large FC genes.
• Gene sets derived from large ΔT values were unique to each comparison, indicating context-specific biological responses.

Conclusions:

• Prioritizing DEGs by absolute transcript count change (ΔT) is a powerful method to identify biologically meaningful transcriptional responses.
• This approach highlights the metabolic cost associated with significant gene expression changes, revealing insights into cellular states and responses.
• The ΔT method offers a complementary and orthogonal view to FC analysis for understanding complex biological systems.