Trajectory & Cell-Fate Agent
Omi's specialised co-pilot for trajectory & cell-fate work
I reconstruct developmental trajectories, infer pseudotime, and predict cell fate decisions from your single-cell data. From a snapshot of cells, I tell you the story of how they got there and where they're heading next.
What I can do for you
I run scVelo, CellRank 2, Palantir, Monocle3, PAGA, or Slingshot depending on what your data supports — and I'll tell you upfront which method is appropriate for your biology vs which would be cargo-culting.
I compute RNA velocity (when you have spliced/unspliced counts), build fate maps with CellRank kernels, and identify driver genes for each lineage — the TFs and surface markers that commit a progenitor down one path vs another.
I detect branching points, quantify lineage priming in early progenitors, and align trajectories across conditions so you can ask: 'is differentiation faster or stalled in the knockout vs control?'
I make the publication-quality trajectory plots that are otherwise a 200-line matplotlib nightmare — stream plots, force-directed layouts, gene-along-pseudotime heatmaps, the works.
Examples of what you can ask me
Copy any of these straight into the demo, or adapt them to your data.
- 1"Run RNA velocity on my hematopoietic dataset."
- 2"Build a CellRank fate map and identify driver genes for each lineage."
- 3"When do CD8 T cells commit to the exhausted vs effector fate?"
- 4"Show me genes changing along the macrophage M1-to-M2 trajectory."
- 5"Compare differentiation pseudotime between control and knockout."
- 6"Find branching points in my organoid dataset."
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
Developmental biologists, stem cell labs, hematopoiesis and immunology researchers tracking differentiation, and anyone working with time-series or organoid datasets.
References
- scVelo (Bergen et al., 2020) – Nature Biotechnology
- CellRank 2 (Weiler et al., 2024) – Nature Methods
- Palantir (Setty et al., 2019) – Nature Biotechnology
- Monocle3 (Cao et al., 2019) – Nature
- PAGA (Wolf et al., 2019) – Genome Biology
- Slingshot (Street et al., 2018) – BMC Genomics
Try Trajectory & Cell-Fate now
Jump into the demo with a starter prompt already loaded. Upload your data, or play with our example dataset first.