Both data are on the same spatial transcriptome slice

[GLking123]

Hi:

The two data are two different species, so the single cell data are different, how about the possibility of de-matching the two different single cell transcriptome data with the data on the spatial transcriptome and plotting the large integrated cell type reference？

Is it separated before it's merged?

Or are they combined and counted together?

For the above question, could you provide some suggestions? Thank you for your valuable time and assistance. I sincerely look forward to your response！

[vitkl]

Hi @GLking123

I would separate the two sections from different species and analyse them separately.

I think you could also use the 1:1 ortholog genes and the integrated reference (single set of cell type labels for both species).

The former is easier. Do you need to compare the species to each other?

[GLking123]

I want to look at them together to see what kind of cell-to-cell communication is happening, because it's a parasitic relationship and one slice will have two species.

[vitkl]

That's a unique data indeed. In principle you can use reference signatures for cell types from both species that have a union of genes from both species. You could estimate the signatures using the reference model for each species separately and then concatenate. You might have to normalise the differences between species (eg scale signatures to have similar average total value mean_f sum_g g_gf in both species) - but that needs to be optimized with some trial and error. Would be nice to know if that works!

[GLking123]

Okay, I will try my best to give you an answer and ask for help in a timely manner.

[GLking123]

We have run cell2location on the same slice using single-cell data from two different species. Is there a way to directly merge the two mapped cell2location data sets? Is there anything wrong with doing this?

[vitkl]

You can concatenate outputs - it could be relatively straightforward for cell abundance in obsm if you have the same set of locations in both analyses. If not it could be more complex but you can

On Sat, 19 Oct 2024 at 16:00, GLking123 @.***> wrote:

We have run cell2location on the same slice using single-cell data from two different species. Is there a way to directly merge the two mapped cell2location data sets?

— Reply to this email directly, view it on GitHub https://github.com/BayraktarLab/cell2location/issues/389#issuecomment-2423936825, or unsubscribe https://github.com/notifications/unsubscribe-auth/AFMFTVYW55HG6XWQG6ZWW3LZ4JXZBAVCNFSM6AAAAABPYNL5QWVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDIMRTHEZTMOBSGU . You are receiving this because you commented.Message ID: @.***>

[GLking123]

@vitkl Hello, the message got cut off.