diffGEK: differential gene expression kinetics.

Area of Science:

• Single-cell RNA sequencing analysis
• Computational biology
• Gene expression dynamics

Background:

• Cellular identity arises from distinct gene expression kinetics, including transcription, splicing, and degradation rates.
• Current RNA velocity models often oversimplify or overparameterize these kinetic rates, limiting their comparative power.
• Accurate estimation of transcriptional kinetics is crucial for understanding cell state transitions.

Purpose of the Study:

• To develop a novel computational method (diffGEK) for comparing transcriptional, splicing, and degradation rates across different biological conditions.
• To overcome limitations of existing RNA velocity models by allowing rates to vary smoothly over a differentiation trajectory.
• To identify genes with altered kinetic rates that may be missed by conventional expression analysis.

Main Methods:

• Developed diffGEK, a method that models transcriptional rates as smooth functions of differentiation.
• Applied diffGEK to analyze erythropoiesis in Jak2 V617F mutant versus wild-type mice.
• Applied diffGEK to analyze myelopoiesis in Ezh2 knockout versus wild-type mice.

Main Results:

• Identified specific genes with altered transcription, splicing, or degradation rates between compared conditions.
• Observed compensatory changes in different kinetic rates that can mask dynamic expression changes in standard analyses.
• Demonstrated the utility of diffGEK in revealing mechanistic differences in gene expression kinetics.

Conclusions:

• diffGEK provides a robust pipeline for comparative expression analysis based on transcriptional kinetics.
• This method uncovers mechanistic differences missed by conventional approaches.
• The approach has broad applicability for analyzing single-cell expression data across various biomedical research questions.