search

mckellardw / scMuscle

The Cornell Single-Cell Muscle Project (scMuscle) aims to collect, analyze and provide to the research community skeletal muscle transcriptomic data
18 stars 4 forks source link

adding inferred cell types #6

Closed skim245 closed 2 years ago

skim245 commented 2 years ago

I have two additional questions related to Fig.4e

are there any easy ways to add cell type predictions to original seurat object? and how can we overlay the inferred cell type content to Visium spot like Fig.4e?

Thank you again :)

mckellardw commented 2 years ago

When you say "original seurat object", do you mean the visium objects or the single-cell objects? In my code, I add the BayesPrism outputs as a separate assay, but you could also add them as columns in the meta.data.

I added the spatial locations of each spot as a separate reduction in the seurat objects (called "space" in my R scripts), then visualized everything with the DimPlot() function from Seurat. This can distort the images if you aren't careful though, so another option would be to follow visualization suggestions from the Seurat team which you can find in this vignette.

skim245 commented 2 years ago

@mckellardw thank you for the prompt response! I want to add the BayesPrism outputs as columns in the VIsium object meta.data. I will try both of your suggestions and see how it goes! Thank you:)