omicverse-single-cell-cellphonedb-communication

What problem does it solve?

This skill turns a CellPhoneDB notebook workflow into a reusable, production-ready tool that runs ligand-receptor statistical analysis on an annotated single-cell AnnData and produces a visualization-ready interaction object for downstream summary plots and pathway analyses. It removes the manual glue work of preparing counts and metadata, running permutations, formatting results for plotting, and choosing among aggregated network, pathway, bubble, chord, and signaling-role visualizations.

Core Features & Use Cases

• Validated CellPhoneDB execution: filters cell types, checks gene-symbol conventions, manages temporary files, and runs permutation-based statistical analysis via ov.single.run_cellphonedb_v5.
• Visualization-ready output: converts results into an AnnData with means and pvalues layers and sender/receiver observation metadata suitable for CellChat-style plotting.
• Branch-rich downstream analysis: compute aggregated sender-receiver networks, pathway-level communication (mean/sum/max/median), extract enriched ligand-receptor pairs, render bubble and chord plots, and compute signaling-role centrality and heatmaps.
• Use case: convert an annotated human scRNA-seq AnnData into a publication-ready communication summary highlighting key sender cell types, pathways, and contributing ligand-receptor pairs.

Quick Start

Run a CellPhoneDB analysis on an annotated AnnData to produce a visualization-ready AnnData and generate aggregated network and pathway plots using ov.single.run_cellphonedb_v5 followed by ov.pl.CellChatViz.