Quentin Blampey

I had a quick look, and I actually have this information in the AnnData object (i.e., `sdata["table"].obs["original_cell_id"]`). Do you confirm you don't have this information? Your screenshot only shows the...

I downgraded to `proseg==2.0.6` and still see the original cell id. Here is the code I used to test it. Can you run this code and confirm that you don't...

Oh I see, sorry for the misunderstanding Sopa creates a new cell id for CosMX data (this is because, for some AtoMx versions, we may not have a global cell...

Thanks for sharing these files @ccruizm When using `sopa.io.cosmx("/path/to/data", cells_table=True)`, the table index should contain the numeric cell IDs, so using `sdata["table"].obs["cell"]` as mentioned above should give you the mapping...

Hello @austinv11, indeed the WSI reader is assuming the images are in RGB, but the `ome_tif` reader tries to look for the correct channel name. I imagine you got a...

Hi @austinv11, I just wanted to clarify the status of this issue: is your solution working? Were you able to read the channel names once you converted the image to...

Okay, I see, makes sense. I'll keep the issue open in case we find a cleaner solution

Hello @derrik-gratz, I would be happy to create a new reader for this technology! Yes, please, if you have details about the output and/or data I could download, it would...

Hi @revalescente, Thanks for reporting, indeed the `no_overlap` feature is still experimental, so it's great to report this kind of bug to help me improve it! I suspect that during...

If possible, all the boundaries of one sample that failed would be perfect, this way I would be able to reproduce it. But it's too complicated, don't worry, I can...