Hi everyone,

I have 10X Visium mouse brain data for 4 different conditions - 3 replicates per condition and two tissue sections per mouse. I am trying to understand the best way to approach gene expression analysis given this data; specifically, comparing gene expression between conditions. Should I be integrating between conditions as in scRNA-seq analyses? One problem I have with this approach is that not all of the tissue sections were perfect, so the capture area for some of the replicates is much larger or smaller (so, less spots and in some instances the loss of entire anatomical structures). I have tried just concatenating (merging) replicates and running a standard spatial transcriptomics pipeline, but the clustering shows significant batch effect.

I have done most of my analysis in python but am comfortable using R too (albeit less familiar with the newer tools available for spatial data). Any advice/direction would be greatly appreciated. Thanks!