Pathway Interpretation Agent
Omi's specialised co-pilot for pathway interpretation work
I take your gene lists, DE results, or cluster markers and turn them into actual biological meaning. Pathway enrichment, gene set scoring, transcription factor activity, and a plain-English explanation of what's going on — no more squinting at GO term lists hoping for inspiration.
What I can do for you
I run enrichment with Enrichr, gseapy, decoupleR, or fgsea against the databases that matter for your biology — Reactome, Hallmark, KEGG, WikiPathways, GO, MSigDB — and consolidate redundant terms so you don't get 47 variations of 'immune response'.
I score every cell for pathway activity (PROGENy, AUCell, UCell, ssGSEA), so you can color a UMAP by 'IFN signalling' or 'glycolysis' instead of guessing from individual genes.
I infer transcription factor activity from your data with decoupleR or SCENIC, and tie it back to expression — so when I say 'STAT1 is active in your inflammatory macrophages', I can show you the target genes that prove it.
I write the biology paragraph for you — what the enriched pathways mean, how they connect to each other, and what the likely upstream regulators are. Saves you an afternoon of reading review articles.
Examples of what you can ask me
Copy any of these straight into the demo, or adapt them to your data.
- 1"Run pathway enrichment on my top 200 cluster markers."
- 2"Score my cells for IFN-α and IFN-γ signatures."
- 3"Which transcription factors are most active in my exhausted T cells?"
- 4"Compare hallmark pathway activity between responders and non-responders."
- 5"Explain the biology behind these 50 differentially expressed genes."
- 6"Run SCENIC and find regulons specific to my regulatory T cells."
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
Biologists who'd rather think about mechanism than mess with database APIs, anyone preparing figures for a paper, and translational researchers connecting transcriptomics to known pathways and drug targets.
References
- Enrichr (Kuleshov et al., 2016) – Nucleic Acids Research
- gseapy (Fang et al., 2023) – Bioinformatics
- decoupleR (Badia-i-Mompel et al., 2022) – Bioinformatics Advances
- fgsea (Korotkevich et al., 2021) – bioRxiv
- PROGENy (Schubert et al., 2018) – Nature Communications
- AUCell / SCENIC (Aibar et al., 2017) – Nature Methods
- MSigDB & Hallmark (Liberzon et al., 2015) – Cell Systems
Try Pathway Interpretation now
Jump into the demo with a starter prompt already loaded. Upload your data, or play with our example dataset first.