Distilling Direct Effects via Conditional Differential Gene Expression Analysis.

Conditional differential gene expression analysis reveals that only a small fraction of genes have direct effects on traits in RNA sequencing data.

• Why it matters: Distinguishing direct effects from correlated or mediated gene associations is crucial for accurate interpretation of gene expression studies and for guiding functional and therapeutic research.
• What they did: The authors developed the GhostKnockoff-based framework called "KnockoffCDGE," which tests for conditional associations using summary statistics from existing DGE pipelines, controlling false discovery rate and handling batch effects, across three datasets.
• The result: Genes identified by CDGE mediate most other differentially expressed genes and show stronger enrichment for known biological interactions than traditional DGE genes, highlighting the importance of focusing on direct effects for downstream applications.