scMTNI: Leveraging cellular trajectory and context to infer dynamic GRNs from single-cell multi-omics data

Area of Science:

• Computational Biology
• Genomics
• Systems Biology

Background:

• Gene regulatory networks (GRNs) are crucial for cell-type-specific gene expression.
• Single-cell multi-omics technologies like scRNA-seq and scATAC-seq provide high-resolution data.
• Existing tools for inferring cell-type-specific GRNs and their dynamics are limited.

Purpose of the Study:

• To develop a computational framework for inferring and analyzing cell-type-specific GRNs.
• To model the dynamics of GRNs in contexts like cellular development and disease progression.
• To provide a comprehensive package for defining and examining GRN dynamics.

Main Methods:

• Developed a multi-task learning framework named single-cell Multi-Task Network Inference (scMTNI).
• Applied scMTNI to a single-cell multi-modal dataset from a cellular reprogramming experiment.
• Utilized scMTNI for inferring cell-type-specific GRNs and identifying key regulators.

Main Results:

• scMTNI successfully infers cell-type-specific GRNs from single-cell multi-omics data.
• The framework enables the analysis of GRN dynamics during cellular transitions.
• Key regulators of cellular fate transitions were identified in the reprogramming dataset.

Conclusions:

• scMTNI is a valuable tool for defining and analyzing cell-type-specific GRNs.
• The framework facilitates the study of dynamic gene regulation in complex biological processes.
• scMTNI aids in understanding cellular fate decisions during development and disease.