Comparative analysis of cell-cell communication at single-cell resolution.

Area of Science:

• Computational Biology
• Genomics
• Cellular Biology

Background:

• Single-cell RNA sequencing (scRNA-seq) enables inference of cell-cell communication (CCC).
• Existing CCC methods analyze data at the cell type or cluster level, losing single-cell resolution.
• There is a need for methods that leverage single-cell information for more precise CCC analysis.

Approach:

• Developed Scriabin, a flexible and scalable framework for comparative CCC analysis at single-cell resolution.
• Validated Scriabin using multiple published datasets, spatial transcriptomics, genetic perturbation screens, and experimental manipulation.
• Applied Scriabin to atlas-scale datasets and longitudinal data to uncover communication pathways and dynamics.

Key Points:

• Scriabin recovers known CCC edges and identifies biologically meaningful interactions.
• The framework reveals co-expressed CCC programs and species-specific communication pathways.
• Scriabin networks track communication dynamics in longitudinal datasets, identifying bystander cells in SARS-CoV-2 infection.

Conclusions:

• Scriabin maximizes the utility of single-cell resolution data for CCC analysis.
• The approach uncovers complex CCC circuitry and niche-phenotype relationships.
• Scriabin offers a broadly applicable strategy for understanding biological systems in health and disease.