CCPLS

CCPLS quantifies how neighboring cell types influence the expression variability of highly variable genes (HVGs) using spatial transcriptome data.

Key Features:

• Focus on Highly Variable Genes (HVGs): Targets HVGs to investigate how cell-cell communications contribute to gene expression variability at single-cell resolution.
• Quantitative Analysis: Utilizes Partial Least Squares (PLS) regression modeling to quantitatively evaluate the influence of multiple neighboring cell types on HVGs.
• Spatial Context Integration: Integrates spatial transcriptome data to link gene expression variability with cellular spatial context within tissues.
• Biological Interpretability: Provides biologically interpretable outputs that describe how neighboring cell types associate with HVG variability in real datasets.
• Statistical Framework: Performs PLS regression modeling for each cell type and reports regression coefficients as quantitative indices of communication effects.

Scientific Applications:

• Developmental Biology: Reveals how neighboring cells influence gene expression variability relevant to mechanisms of normal development.
• Disease Research: Characterizes altered cell-cell communications in disease contexts by analyzing spatial transcriptome-derived HVG variability.

Methodology:

Partial Least Squares (PLS) regression modeling is applied per cell type to quantify effects of multiple neighboring cell types on HVGs in spatial transcriptome datasets, with regression coefficients reported as quantitative indices of communication effects.