Single-cell CellPhoneDB communication mapping

Ligand-receptor mapping quantifies cell-cell communication by identifying receptor-ligand pairs between annotated cell types. This Skill runs CellPhoneDB v5 on your annotated single-cell dataset to infer interactions, calculate interaction strength, and output networks ready for visualization. Input requires log-normalized AnnData with cell type labels; output includes interaction matrices and CellChat-style plots.

You need a filtered, log-normalized AnnData object with cell type annotations in a 'cell_labels' column. CellPhoneDB requires omicverse, scanpy, anndata, pandas, matplotlib, and the CellPhoneDB v5 bundle installed. The Skill handles interaction inference and generates cpdb_results and visualization-ready output.

Yes. This Skill generates CellChat-style visualizations including circle, chord, and heatmap plots that show aggregated communication strength and pathway-specific signaling between cell types. Output includes global network views and sender-receiver role analysis across your annotated populations.

CellPhoneDB works with any annotated single-cell dataset where cells are grouped into labeled populations. This Skill is designed for datasets with clear cell type annotations and log-normalized expression. It outputs interaction networks and visualizations specific to your cell types and significant signaling pathways.

CellPhoneDB infers potential interactions from expression patterns; it does not measure direct binding or validate functional communication. Results depend on annotation quality and expression filtering. The method works best with sufficient cells per type and detectable ligand-receptor expression across populations.

This Skill quantifies communication strength across all ligand-receptor pairs and summarizes pathway-specific interactions. Output includes ranked interaction networks and visualizations highlighting significant pathways, enabling you to prioritize biological hypotheses for downstream validation.