A cell type and state specific gene regulation network inference method for immune regulatory analysis.

Area of Science:

• Genomics
• Computational Biology
• Systems Biology

Background:

• Bulk sequencing struggles to resolve cell-type and state-specific gene regulatory networks.
• Single-cell RNA sequencing (scRNA-seq) offers potential for detailed network inference.
• Existing methods overlook dynamic network changes and cell-type heterogeneity.

Purpose of the Study:

• To develop a method for inferring cell type and state-specific gene regulatory networks (CSNs) using scRNA-seq data.
• To address limitations of bulk sequencing and improve network inference accuracy.
• To provide a robust tool for analyzing complex biological systems.

Main Methods:

• Inferring pseudo-temporal information and reordering cells.
• Windowing cells based on state to mitigate density-related biases.
• Employing sparse regression with reference network data for CSN construction.

Main Results:

• inferCSN demonstrates superior performance over existing methods on simulated and real scRNA-seq data.
• The method shows robustness across datasets of varying types and scales.
• Analysis of T cell regulatory networks revealed immune suppression pathways.

Conclusions:

• inferCSN accurately infers cell type and state-specific gene regulatory networks.
• The method is robust and applicable to diverse scRNA-seq datasets.
• Comparative network analysis can uncover critical biological pathways, such as immune suppression.