Advancing spatial cellular communication inference with ligand diffusion and transport model.

Area of Science:

• Molecular Biology
• Systems Biology
• Bioinformatics

Background:

• Cell-cell communication is crucial for multicellular organisms, relying on ligand-receptor interactions.
• Traditional methods for studying cellular communication lack spatial resolution and are prone to errors.
• Spatial transcriptomics enables analysis of communication events with precise spatial information.

Purpose of the Study:

• To develop a novel computational framework for inferring spatial cellular communication at single-cell resolution.
• To integrate ligand diffusion, competitive binding, and transport dynamics into a unified model.
• To predict downstream target genes regulated by ligand-receptor interactions.

Main Methods:

• Introduced SCILD (Spatial Cellular communication Inference with Ligand Diffusion and transport model), an optimization-based framework.
• Modeled cellular communication as a cargo transport system with potential losses, incorporating ligand diffusion and competitive binding.
• Utilized neural network modeling and in silico perturbation for target gene prediction.

Main Results:

• SCILD accurately captures competitive communication dynamics at the single-cell level.
• Identified biologically relevant ligand-receptor markers driving domain-specific signaling.
• Resolved subdomain-specific communication patterns and robustly predicted target genes.
• Validated findings against external databases.

Conclusions:

• SCILD is a versatile and powerful tool for spatial cellular communication research.
• The framework enhances the understanding of cell-cell signaling in complex biological systems.
• SCILD provides high-resolution insights into communication dynamics previously unattainable.