scSAGRN: Inferring gene regulatory networks from single-cell multi-omics using spatial association

Area of Science:

• Genomics
• Systems Biology
• Computational Biology

Background:

• Understanding gene regulatory relationships is crucial for deciphering biological processes and diseases.
• Single-cell multi-omics data offers a powerful lens for studying transcriptional regulation at an unprecedented resolution.
• Resolving connections between cis-regulatory elements and genes is key to mapping these networks.

Purpose of the Study:

• To introduce scSAGRN, a novel computational framework for inferring gene regulatory networks (GRNs) from single-cell multi-omics data.
• To enhance the prediction of transcription factor (TF) and target gene interactions.
• To provide a robust method for identifying key TFs and regulatory elements within complex biological systems.

Main Methods:

• scSAGRN integrates spatial association to correlate gene expression with chromatin accessibility data.
• The framework links distal cis-regulatory elements (e.g., enhancers) to their target genes.
• It infers GRNs and identifies critical transcription factors using a multi-omics approach.

Main Results:

• scSAGRN demonstrates superior performance in TF recovery and peak-gene/TF-gene linkage prediction compared to existing methods.
• The framework successfully inferred GRNs and identified key TFs in human peripheral blood mononuclear cells and mouse brain datasets.
• Benchmarking on real single-cell datasets validates the efficacy and robustness of scSAGRN.

Conclusions:

• scSAGRN provides a significant advancement in predicting transcriptional regulatory patterns from single-cell multi-omics data.
• The framework offers a valuable tool for researchers studying gene regulation in development and disease.
• It establishes a new reference for dissecting complex gene regulatory networks at single-cell resolution.