Cell-Cell Communication Agent
Omi's specialised co-pilot for cell-cell communication work
I infer ligand-receptor signalling between your cell types — who's sending what message to whom, which pathways light up in disease vs healthy, and how communication rewires across conditions. The cellular gossip column, but rigorously inferred.
What I can do for you
I run CellChat, CellPhoneDB, NicheNet, or LIANA on your dataset and aggregate the results so you don't have to puzzle over conflicting outputs from five tools — I'll explain where they agree and where they don't.
I rank the strongest interactions per cell-type pair, surface upstream-vs-downstream relationships (NicheNet), and identify which signalling pathways (TGFβ, WNT, IFN, etc.) dominate your tissue.
I compare communication networks between conditions: what new circuits open up in tumor vs adjacent normal? Which interactions are lost in old vs young? Beautiful chord diagrams and heatmaps included.
I link signalling predictions back to gene expression and (if you have it) spatial coordinates — because a predicted interaction is much more credible when sender and receiver are actually next to each other.
Examples of what you can ask me
Copy any of these straight into the demo, or adapt them to your data.
- 1"Which cell types are signalling to my exhausted CD8 T cells?"
- 2"Compare cell-cell communication in tumor vs adjacent normal tissue."
- 3"Run NicheNet: which ligands from macrophages drive my fibroblast activation?"
- 4"Show me the TGFβ signalling network in my dataset."
- 5"Which interactions are gained in disease and lost in healthy?"
- 6"Make a chord diagram of the top 30 interactions between my clusters."
How I work
I run real Scanpy (Python) or Seurat (R) code on the secure MCP server — no hallucinations, no made-up gene lists. Every result comes with the exact code I executed and the parameters I used, so your analysis is fully reproducible and ready for the Methods section.
Best for
Tumor microenvironment researchers, developmental biologists, immunologists studying tissue niches, and anyone who wants mechanistic hypotheses for downstream wet-lab validation.
References
- CellChat (Jin et al., 2021) – Nature Communications
- CellPhoneDB (Efremova et al., 2020) – Nature Protocols
- NicheNet (Browaeys et al., 2020) – Nature Methods
- LIANA+ (Dimitrov et al., 2024) – Nature Cell Biology
Try Cell-Cell Communication now
Jump into the demo with a starter prompt already loaded. Upload your data, or play with our example dataset first.