CCPLS reveals cell-type-specific spatial dependence of transcriptomes in single cells.

Area of Science:

• * Spatial transcriptomics and computational biology.
• * Molecular mechanisms of cell-cell communication.

Background:

• * Cell-cell communications are vital for cellular states, development, and disease.
• * Single-cell RNA sequencing reveals cell-to-cell expression variability in highly variable genes (HVGs).
• * Understanding how cell-cell communications regulate HVG variability is crucial but underexplored, especially with spatial context.

Purpose of the Study:

• * To develop a quantitative and interpretable computational framework for analyzing cell-cell communications.
• * To specifically investigate the impact of multiple neighboring cell types on the cell-to-cell expression variability of HVGs.
• * To leverage spatial transcriptomics data for uncovering these regulatory relationships.

Main Methods:

• * Proposed CCPLS (Cell-Cell communications analysis by Partial Least Square regression modeling), a statistical framework.
• * Utilized partial least squares (PLS) regression modeling for each cell type.
• * Quantified cell-cell communications using regression coefficients as indices.

Main Results:

• * CCPLS accurately estimated the effects of multiple neighboring cell types on HVGs in simulated data.
• * Applied to real datasets, CCPLS extracted biologically interpretable insights.
• * Demonstrated the method's ability to handle complex multi-neighbor interactions.

Conclusions:

• * CCPLS provides a robust method for analyzing cell-cell communications impacting HVG variability.
• * The framework enhances the interpretability of spatial transcriptomics data.
• * Offers a novel approach to understanding gene expression regulation in complex cellular environments.