Distinguish spot-based vs single-cell platforms for annotation/deconv…#2
Distinguish spot-based vs single-cell platforms for annotation/deconv…#2
Conversation
…olution routing Add platform-aware guidance so the agent correctly routes spot-based data (Visium, Slide-seq, ST) to spatial_deconvolution and single-cell resolution data (MERFISH, Xenium, CosMx, SeqFISH) to annotation.
|
thx! I agree that two modalities could benefit from more tailiored context. Could you add some notebooks in the notebooks folder under the branch https://github.com/Genentech/SpatialAgent/tree/case I want to see how the agent perform on the two types of data after updating the skills. |
hansen7
left a comment
hansen7
left a comment
There was a problem hiding this comment.
provide use cases (annotation on two kinds of spatial data)
Demonstrates platform-aware routing: Visium spot-based data correctly routes to spatial_deconvolution (Stereoscope) instead of single-cell annotation. Notebook includes full agent run with 13 cell types identified and spatial visualizations.
|
@hansen7 Add Visium cell type annotation demo notebook with executed outputs |
hansen7
left a comment
hansen7
left a comment
There was a problem hiding this comment.
thx! can you push the notebook to the https://github.com/Genentech/SpatialAgent/tree/case branch, then I'll merge other changes.
|
added to case branch |
hansen7
left a comment
hansen7
left a comment
There was a problem hiding this comment.
just remove the notebook in the current pr, then it should be ready to go!
|
Removed the notebook from this PR. It's already on the case branch. |
2 participants
…olution routing
Add platform-aware guidance so the agent correctly routes spot-based data (Visium, Slide-seq, ST) to spatial_deconvolution and single-cell resolution data (MERFISH, Xenium, CosMx, SeqFISH) to annotation.