Comparative Atlas Mapper Agent
Omi's specialised co-pilot for atlas mapper work
I map your dataset against the great public reference atlases — CELLxGENE, Tabula Sapiens, Human Cell Atlas, Allen Brain Atlas, Tabula Muris — so you can borrow their annotations, validate your cell types, and place your data in global context. Standing on the shoulders of giants, conversationally.
What I can do for you
I query CELLxGENE Census, Tabula Sapiens, HCA, Allen, and the Human Lung/Heart/Gut/Tumor Atlases directly, pull matched reference subsets, and integrate them with your data using scArches, scANVI, or Symphony — preserving the reference labels.
I label-transfer cell type annotations from the reference to your cells with calibrated confidence scores, so you know which calls to trust and which need a second look.
I flag novel populations in your data — cells that don't match anything in the reference — and surface their distinguishing markers, because that's where your paper's discoveries usually live.
I quantify how your dataset compares to the reference: cell-type proportions, expression shifts within matched populations, and disease-vs-healthy contrasts using the reference as the 'normal' baseline.
Examples of what you can ask me
Copy any of these straight into the demo, or adapt them to your data.
- 1"Map my PBMC dataset to Tabula Sapiens and transfer cell type labels."
- 2"Compare my tumor cells to the Human Tumor Atlas."
- 3"Find populations in my data that aren't in CELLxGENE."
- 4"Annotate my brain dataset using the Allen Brain Atlas reference."
- 5"How do my macrophages compare to the lung atlas reference macrophages?"
- 6"Pull a matched healthy reference from CELLxGENE for my disease samples."
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
Anyone working in a well-mapped tissue (immune, brain, lung, gut, heart), labs without their own healthy controls, and atlas-builders contributing back to public resources.
References
- CELLxGENE Census (CZI Single-Cell Biology Program, 2023)
- Tabula Sapiens (The Tabula Sapiens Consortium, 2022) – Science
- Human Cell Atlas (Regev et al., 2017) – eLife
- scArches (Lotfollahi et al., 2022) – Nature Biotechnology
- Symphony (Kang et al., 2021) – Nature Communications
- Allen Brain Atlas (Lein et al., 2007) – Nature
Try Atlas Mapper now
Jump into the demo with a starter prompt already loaded. Upload your data, or play with our example dataset first.