NetREm: Network Regression Embeddings reveal cell-type transcription factor coordination for gene regulation

Area of Science:

• Computational Biology
• Genomics
• Systems Biology

Background:

• Transcription factor (TF) coordination is crucial for gene regulation through protein-protein interactions (PPIs) and DNA co-binding.
• Single-cell technologies enable gene expression analysis but understanding TF coordination in diverse cell types remains a challenge.

Purpose of the Study:

• To introduce Network Regression Embeddings (NetREm), a novel computational approach to identify cell-type-specific TF-TF coordination and target gene (TG) regulation.
• To uncover cell-type coordinating TFs and predict novel TF-TG regulatory links using single-cell gene expression data.

Main Methods:

• NetREm utilizes network-constrained regularization, incorporating prior PPI knowledge.
• Analyzes single-cell gene expression data to infer TF coordination and regulatory networks.
• Validated through simulation studies and benchmarking across human, mouse, and yeast datasets.

Main Results:

• NetREm successfully identified cell-type TF-TF coordination and TF-TG regulatory links.
• Prioritized novel human TF-TF coordination links in immune cell subtypes.
• Revealed cell-type networks in nervous systems and in Alzheimer's disease, with top predictions experimentally validated.

Conclusions:

• NetREm is a powerful tool for dissecting TF coordination and gene regulation across cell types.
• The approach facilitates the discovery of novel regulatory relationships and links genetic variants to regulatory networks.
• Provides insights into TF networks in complex diseases like Alzheimer's and across different tissue types.