Context-dependent gene regulatory network reveals regulation dynamics and cell trajectories using unspliced...

Area of Science:

• Molecular Biology
• Systems Biology
• Genomics

Background:

• Gene regulatory networks (GRNs) are crucial for biological processes like development and disease.
• Single-cell RNA sequencing (scRNA-seq) offers high resolution but challenges exist in inferring comprehensive GRNs.
• Existing methods struggle to capture cell-type-specific regulatory dynamics.

Purpose of the Study:

• To develop a novel method for constructing context-dependent gene regulatory networks (CDGRNs) from scRNA-seq data.
• To leverage both spliced and unspliced transcripts for more accurate network inference.
• To link molecular regulatory mechanisms to macroscopic cell differentiation processes.

Main Methods:

• Utilized scRNA-seq data, incorporating both spliced and unspliced transcript levels.
• Decomposed GRNs into context-specific subnetworks using Gaussian mixture models.
• Inferred consensus transcription factor-target gene regulatory pairs within each subnetwork.

Main Results:

• The union of inferred regulatory pairs across all contexts successfully reconstructed cell differentiation trajectories.
• Context-specific subnetworks showed enrichment for cell cycle, differentiation, and tissue-specific functions.
• Network entropy of CDGRNs decreased along differentiation trajectories, indicating directed developmental processes.

Conclusions:

• CDGRNs provide a powerful framework for understanding gene regulation in diverse cell types.
• This approach connects molecular-level gene regulation to cell differentiation dynamics.
• CDGRNs offer insights into the directionality and mechanisms of developmental processes.