Deconvolute spatial transcriptomics (visium) with unannotated single-cell (10x)

General Visium
1

I have a spatial transcriptomics dataset (10x visium) of a tumour microarray and some of the samples have matched single-cell RNA-seq data (10x). I would like to annotate my spatial data with the single-cell data. However the single-cell data is not annotated. Can this be done with any scverse tool?

2

Not sure I followed.
Do you have any annotations at all? whether on the spatial data or the single cell data?

Basically you can use scvi-tools for annotate your data as a query data but only given you have some suitable reference data which is annotated (scvi/scanvi).
Following that you can use other tools to also try to annotate that overlapping part of your spatial data.
But you can perhaps auto-annotate your spatial data using tools like cell2location.

3

Hi @ori-kron-wis
The matched single-cell data does not have any annotations and is just processed to GEX count matrices. I understand that this needs to be annotated first in order to transfer labels?

Does cell2location annotate spatial data based on a single-cell reference?

Thanks

4

Yes. I suggested that it will be annotated using a model like scanvi (if you have similar annotated data as references) or cellassign (if you know whether or not each given gene is a marker of different cell types you think you have).

Yes. that will be the 2nd part , if your RNA-seq data will be annotated you will be able to use it and estimate the annotation of the spatial part with cell2location based on location of cells.
I guess there are other tools for that.