seuratToGiotto5 doesn't work

[marijazmf]

Hi,
Thank you for the great Giotto package! I'm trying to import a Seurat object into Giotto, but unfortunately, the function doesn’t seem to work. I'm using Giotto Suite version 4.2.2 and GiottoClass version 0.4.8. However, when I try to run seuratToGiotto5(), I encounter an error:

ERROR: Definition of function 'seuratToGiotto5' not found

in namespace 'package:GiottoClass'

The installation completed without any errors, and everything seems to be in order. The function even appears in the documentation. However, when I try to run it, R tells me the function cannot be found.

Thanks,
Marija

[jiajic]

Hi @marijazmf,

Thanks for using Giotto!
I notice a misspelling: can you try seuratToGiottoV5() instead? Notice the V

Best,
George

[marijazmf]

Hi @jiajic,

Thank you for the quick response. The "V5" was indeed a typo, but even after correcting it, I encountered the same error. The issue was ultimately resolved by reinstalling the GiottoClass package.

However, I’m still facing an error when trying to convert my Seurat object to Giotto. I’ve already integrated, normalized, and clustered the Xenium data in Seurat, and I would like to use Giotto’s spatial analysis functionalities without repeating those steps. Specifically, I want to preserve the existing clusters and normalized counts from Seurat during the conversion.

giotto_o <- seuratToGiottoV5(seurat_sample,
+                                spatial_assay = "Xenium",
+                                dim_reduction = c("pca", "umap"),
+                                subcellular_assay = "SCT")
Warning: data layer is not found and counts layer is used
  Selecting col "molecule" as feat_ID column
  Selecting cols "x" and "y" as x and y respectively
python already initialized in this session
 active environment : 'giotto_env'
 python version : 3.8
Error in .check_spatial_location_data(gobject) : 
  Number of rows of spatial locations do not match
                        with cell IDs for: 
 spatial unit: cell and coordinates: raw 

Can you guess what can be the problem, as I've already adjusted the number of genes and the number of molecules, but still dosn't work.

length(rownames(seurat_sample))
[1] 360
> seurat_sample@images$mPso10.resolved
Spatial coordinates for 8674 cells and 360 molecules
 First 10 molecules: ACER1, ACTA2, ADAM12, ADRB2, AHNAK2, AHR
Default segmentation boundary: centroids 
 1 other segmentation boundaries present: segmentation 
Associated assay: Xenium 
Key: Xenium_ 

Thanks,
Marija

[jiajic]

Not too sure about this myself.

The error here is saying that the number of spatial cell centroids that Giotto has found does not match the number of cell_IDs in the expression information or polygons.

@iqraAmin, do you have any ideas about what might be going wrong?

[iqraAmin]

@marijazmf I've tested it on my data and it appears to work correctly. Could you provide a small subset of your data so I can investigate what's causing the error?

[marijazmf]

Dear @jiajic and @iqraAmin,

thank you both for your quick support. Here is one sample that didn't work.
Looking forward to your answer.
Best,
Marija

[iqraAmin]

Hi @marijazmf, we've updated the Seurat interoperability function. Please try reinstalling GiottoClass and using the function, and let us know if it works for you now.
Best,

[marijazmf]

Thanks @iqraAmin, it works now!

best,
Marija