Skip to content

usage-guide.md

Latest commit

 

History

History
49 lines (32 loc) · 1.33 KB

usage-guide.md

File metadata and controls

49 lines (32 loc) · 1.33 KB
Raw

Spatial Cell-Cell Communication - Usage Guide

Overview

This skill covers analyzing cell-cell communication in spatial transcriptomics data using ligand-receptor analysis with Squidpy. Identify which cell types are communicating and through which signaling pathways.

Prerequisites

pip install squidpy scanpy pandas networkx matplotlib

Quick Start

Tell your AI agent what you want to do:

  • "Run ligand-receptor analysis on my spatial data"
  • "Find which cell types are communicating"

Example Prompts

Basic Analysis

"Analyze cell-cell communication in my Visium data"

"Run ligand-receptor analysis between cell types"

Specific Interactions

"Find interactions between T cells and macrophages"

"Check if CCL2-CCR2 signaling is active"

Visualization

"Show a heatmap of cell-cell interactions"

"Plot the communication network"

What the Agent Will Do

  1. Verify cell type annotations exist
  2. Build spatial neighbor graph
  3. Run ligand-receptor permutation test
  4. Filter significant interactions
  5. Visualize results

Tips

  • Cell type annotations - Required before running communication analysis
  • Spatial neighbors - Analysis considers spatial proximity
  • Permutations - More permutations = more robust p-values but slower
  • Custom databases - Can use custom ligand-receptor pair lists