scale.factors or spatial.factors?

[ozcelikelif]

In the tutorial https://htmlpreview.github.io/?https://github.com/jinworks/SpatialCellChat/blob/master/tutorial/SpatialCellChat_analysis_of_spatial_transcriptomics_data.html the cellchat object is created using spatial.factors which includes conversion.factor and tol parameters. This parameter does not exist in the function createCellChat anymore. I am guessing this parameter is updated by scale.factors which now requires other things then ratio and tol, which is what we were putting inside spatial.factors in the tutorials posted.

Error in createCellChat(object = data.input, meta = meta, group.by = "labels", :
scale.factors with elements named spot.diameter and spot should be provided!

Could you please clarify what this change is and now where are we providing the ratio/tol or how are you doing the pixel to micron conversion if not asking for the conversion rate?

[Ethel-0]

@ozcelikelif Since version 2.1.1 of CellChat, we have updated the parameter scale.factors to spatial.factors. Please check your CellChat version. In addition, when creating an object for spatial data using SpatialCellChat, use the function createSpatialCellChat.

Note that CellChat and SpatialCellChat contain similarly named functions, so please make sure you are using the correct package during your analysis.

[ozcelikelif]

Which one would you recommend using for CosMX/Xenium analysis? I am having trouble running the spatial version of cellchat. Also, when you create spatial cell chat object does it limit the ligand-receptor data specifically to the panel of genes available in my spatial object or is that something i have to manually do? Thank you so much!

[Ethel-0]

Which one would you recommend using for CosMX/Xenium analysis? I am having trouble running the spatial version of cellchat. Also, when you create spatial cell chat object does it limit the ligand-receptor data specifically to the panel of genes available in my spatial object or is that something i have to manually do? Thank you so much!

@ozcelikelif 1. Spatial CellChat (i.e., CellChat v3) is better suited for spatial data from CosMX/Xenium analyses, as it incorporates spatial distance constraints to infer interactions at single-cell resolution and provides more visualization approaches specifically designed for spatial data.

2. After creating the Spatial CellChat object, all subsequent analyses are restricted to the genes available in the spatial dataset, and no manual filtering is required. For more details, please refer to the tutorial.