Help Needed: Subsetting Individual Samples from Xenium TMA Data in Seurat

[FenosoaRandrianjatovo]

Hello,

I’m working with Xenium TMA data in Seurat and need to extract each patient’s sample as its own field of view (FOV) and count matrix. When I load my Xenium run with LoadXenium(), I end up with one aggregated Seurat object containing all cores and a single count matrix.

I’ve tried manually cropping the spatial object using Crop(), but the resulting region isn’t aligned with the individual core boundaries.

Could anyone share the best approach to:

      1-Split a TMA run into separate FOVs (one per patient)
      
      2-Obtain the corresponding count matrix for each FOV

Any guidance or example code would be greatly appreciated!

Thank you,
Fenosoa