We'll likely showcase the pseudo-bulked data with k = 9 in the app. That will be for the layer-like level analysis (so differences across the k = 9 clusters). Though we haven't finalized this decision given the region analysis that we are also doing.
This can wait a bit, but I wanted to make this issue before I forget about it ;)
You can see https://github.com/LieberInstitute/Visium_IF_AD/blob/master/code/99_prepare_for_shiny/01_prepare_shiny.R and https://github.com/LieberInstitute/Visium_IF_AD/tree/master/code/05_deploy_app_wholegenome for how we've deployed pseudo-bulked data. Note that the pseudo-bulked data will need something like this https://github.com/LieberInstitute/Visium_IF_AD/blob/3c7a829dc2bee97d8d7850c9df40e2eec6335ef0/code/11_grey_matter_only/01_create_pseudobulk_data.R#L176-L186.
We'll likely showcase the pseudo-bulked data with k = 9 in the app. That will be for the layer-like level analysis (so differences across the k = 9 clusters). Though we haven't finalized this decision given the region analysis that we are also doing.
This can wait a bit, but I wanted to make this issue before I forget about it ;)