How to merge Spatial data

[EZUY]

Hi All,
I'm currently trying to merge multiple spatial data generated with spaceranger count.

I first tried to use aggregated matrix with spaceranger aggr

data_dir<-"Seurat\\Aggr"
A1_10X_Spatial<-Load10X_Spatial(data.dir = data_dir,filename ="filtered_feature_bc_matrix.h5")`

Then I have
Error in readPNG(source = file.path(image.dir, image.name)) :
unable to open Seurat\Aggr/spatial/tissue_lowres_image.png

a1 <- Load10X_Spatial(data.dir="Seurat\\A1", filename = "filtered_feature_bc_matrix.h5", assay = "Spatial", slice="slice1")
b1 <- Load10X_Spatial(data.dir="Seurat\\B1", filename = "filtered_feature_bc_matrix.h5", assay = "Spatial", slice="slice2")
c1 <- Load10X_Spatial(data.dir="Seurat\\C1", filename = "filtered_feature_bc_matrix.h5", assay = "Spatial", slice="slice3")
d1 <- Load10X_Spatial(data.dir="Seurat\\D1", filename = "filtered_feature_bc_matrix.h5", assay = "Spatial", slice="slice4")
a1 <- SCTransform(a1, assay = "Spatial", verbose = FALSE)
b1 <- SCTransform(b1, assay = "Spatial", verbose = FALSE)
c1 <- SCTransform(c1, assay = "Spatial", verbose = FALSE)
d1 <- SCTransform(d1, assay = "Spatial", verbose = FALSE)
a1[["orig.ident"]] <- "a1"
b1[["orig.ident"]] <- "b1"
c1[["orig.ident"]] <- "c1"
d1[["orig.ident"]] <- "d1"
st_seurat_merge <- a1
st_seurat_merge <- merge(st_seurat_merge, a1)
st_seurat_merge <- merge(st_seurat_merge, b1)
st_seurat_merge <- merge(st_seurat_merge, c1)
st_seurat_merge <- merge(st_seurat_merge, d1)

when I'm merging them. I got Warning: Some cell names are duplicated across objects provided. Renaming to enforce unique cell names.Warning: Some cell names are duplicated across objects provided. Renaming to enforce unique cell names.

when I run

st_seurat_merge <- RunPCA(st_seurat_merge, assay = "SCT", verbose = FALSE)
st_seurat_merge <- FindNeighbors(st_seurat_merge, reduction = "pca", dims = 1:30)
st_seurat_merge <- FindClusters(st_seurat_merge, verbose = FALSE)
st_seurat_merge <- RunUMAP(st_seurat_merge, reduction = "pca", dims = 1:30)

I got
Error in PrepDR(object = object, features = features, verbose = verbose) :
Variable features haven't been set. Run FindVariableFeatures() or provide a vector of feature names.

Usually, how do I merge different spatial matrix&picture to one Seurat object?

[EZUY]

nvm. I added

DefaultAssay(st_seurat_merge) <- "SCT"
VariableFeatures(st_seurat_merge) <- c(VariableFeatures(a1), VariableFeatures(b1),VariableFeatures(c1), VariableFeatures(d1))

[EZUY]

May I know how to get the gene expression in different clusters of merged Spatial data?

[Pancreas-Pratik]

Hi @EZUY

I have two VisiumHD datasets (same tissue, different conditions) running through spaceranger count at the moment, but am wondering how to appropriately proceed next.

What did you end up doing?

Did you use the output from spaceranger count only? or did you use any ouputs from spaceranger aggr?

and then what did you do afterwards for downstream analysis? was merging necessary?

Thank you very much in advance.

---

EDIT: I think I found the solution that I will try (which looks like what you did), which leads me to believe that maybe spaceranger aggr output can be used in some way:
https://satijalab.org/seurat/articles/spatial_vignette#working-with-multiple-slices-in-seurat

and also try with a combination of this section for DGE analysis:
https://satijalab.org/seurat/articles/visiumhd_analysis_vignette#unsupervised-clustering

---

EDIT: spaceranger v3.1.1 does not support aggr for VisiumHD