Integrating multiple Visium HD samples

[pratibha-bioinfo]

Hi Seurat team,

Thank you for providing support to analyze Visium HD samples. This is the first time I am using github so apologies if I did not follow syntax.

Is it correct to follow the 10X spatial integration vignette[https://satijalab.org/seurat/articles/spatial_vignette.html#working-with-multiple-slices-in-seurat] to integrate multiple Visium HD samples?

Kindly,
Pratibha

[Scathacheng]

Hi Pratibha and Seurat team,

I am encountering the same problem and showing my problem under the same issue.

I am analyzing 10X Visium HD data of 15 samples. Following the vignette of 10X spatial integration, I found some confusing problems:

1. I used to analyze scRNA-seq data using Seurat V2 and SCTransform, the Seurat object has two assays of "RNA" and "SCT". However, when I tried to analyze 10X Visium HD data (https://satijalab.org/seurat/articles/visiumhd_analysis_vignette), I found that the Seurat object already has two assays of "8 um" and "16 um", which cannot perform “SCTransform” function directly. I want to know if it is ok that I do not run “SCTransform” in integrating multiple slices, as shown in the vignette (https://satijalab.org/seurat/articles/spatial_vignette.html#working-with-multiple-slices-in-seurat).

2. I next used “harmony” to correct the batch effect. However, it also reports a warning that “quick transfer stage steps exceeded maximum”, and the harmony only finished 1 cycle (1/10). I am concerned that the results are not reproducible.

3. Here is my code. Please tell me if there is any problem or if there is anything I did wrong. Much appreciate!

dir = c("01rawdata/04alldata/X1725569","01rawdata/04alldata/X1807783","01rawdata/04alldata/X1913448","01rawdata/04alldata/X2008632","01rawdata/04alldata/X201426205","01rawdata/04alldata/X202031171","01rawdata/04alldata/X202035595","01rawdata/04alldata/X2020586","01rawdata/04alldata/X202239003","01rawdata/04alldata/X202242073","01rawdata/04alldata/X20230970","01rawdata/04alldata/X202410354","01rawdata/04alldata/X2123326","01rawdata/04alldata/X2307031","01rawdata/04alldata/X2313317")

names(dir) = c("X1725569","X1807783","X1913448","X2008632","X201426205","X202031171","X202035595","X2020586","X202239003","X202242073","X20230970","X202410354","X2123326","X2307031","X2313317")

seu_list <- list()
for (i in 1:length(dir)) {
seu_list[[i]] = Load10X_Spatial(data.dir = dir[i])
seu_list[[i]]@meta.data$orig.ident = names(dir)[i]
}

seurat_ob = merge(seu_list[[1]], y=c(seu_list[[2]],seu_list[[3]],seu_list[[4]],seu_list[[5]],seu_list[[6]],seu_list[[7]],seu_list[[8]],seu_list[[9]],seu_list[[10]],seu_list[[11]],seu_list[[12]],seu_list[[13]],seu_list[[14]],seu_list[[15]]))

here, I decide to run the function from “NormalizeData” to “FindClusters” using the 0.16um and 0.08um, separately. I have no idea if this is right.

DefaultAssay(seurat_ob) <- "Spatial.016um"
seurat_ob <- NormalizeData(seurat_ob)
seurat_ob <- FindVariableFeatures(seurat_ob)
seurat_ob <- ScaleData(seurat_ob)
seurat_ob <- RunPCA(seurat_ob, verbose=FALSE)
seurat_ob <- RunHarmony(seurat_ob, group.by.vars = "orig.ident") #some warnings reported in this step
seurat_ob <- RunUMAP(seurat_ob, reduction="harmony", dims=1:30)
seurat_ob <- FindNeighbors(seurat_ob, reduction="harmony", dims=1:30)
seurat_ob <- FindClusters(object = seurat_ob, resolution = 0.4, algorithm = 1, verbose = F)

DefaultAssay(seurat_ob) <- "Spatial.008um"
seurat_ob <- NormalizeData(seurat_ob)
seurat_ob <- FindVariableFeatures(seurat_ob)
seurat_ob <- ScaleData(seurat_ob)
seurat_ob <- RunPCA(seurat_ob, verbose=FALSE)
seurat_ob <- RunHarmony(seurat_ob, group.by.vars = "orig.ident")
seurat_ob <- RunUMAP(seurat_ob, reduction="harmony", dims=1:30)
seurat_ob <- FindNeighbors(seurat_ob, reduction="harmony", dims=1:30)
seurat_ob <- FindClusters(object = seurat_ob, resolution = 0.4, algorithm = 1, verbose = F)

4. Here is my sessionInfo()

R version 4.3.3 (2024-02-29)
Platform: x86_64-conda-linux-gnu (64-bit)
Running under: CentOS Linux 7 (Core)

Matrix products: default
BLAS/LAPACK: /gpfs/hpc/home/zhangyongch/jcdai/anaconda3/envs/myR/lib/libopenblasp-r0.3.21.so; LAPACK version 3.9.0

locale:
[1] LC_CTYPE=en_US.UTF-8 LC_NUMERIC=C
[3] LC_TIME=en_US.UTF-8 LC_COLLATE=en_US.UTF-8
[5] LC_MONETARY=en_US.UTF-8 LC_MESSAGES=en_US.UTF-8
[7] LC_PAPER=en_US.UTF-8 LC_NAME=C
[9] LC_ADDRESS=C LC_TELEPHONE=C
[11] LC_MEASUREMENT=en_US.UTF-8 LC_IDENTIFICATION=C

time zone: Asia/Shanghai
tzcode source: system (glibc)

attached base packages:
[1] stats graphics grDevices utils datasets methods base

other attached packages:
[1] harmony_1.2.3 Rcpp_1.0.13-1 Seurat_5.1.0 SeuratObject_5.0.2
[5] sp_2.1-4

loaded via a namespace (and not attached):
[1] deldir_2.0-4 pbapply_1.7-2 gridExtra_2.3
[4] rlang_1.1.4 magrittr_2.0.3 RcppAnnoy_0.0.22
[7] matrixStats_1.4.1 ggridges_0.5.6 compiler_4.3.3
[10] spatstat.geom_3.3-4 png_0.1-8 vctrs_0.6.5
[13] reshape2_1.4.4 stringr_1.5.1 pkgconfig_2.0.3
[16] fastmap_1.2.0 promises_1.3.2 purrr_1.0.2
[19] jsonlite_1.8.9 goftest_1.2-3 later_1.4.1
[22] spatstat.utils_3.1-1 irlba_2.3.5.1 parallel_4.3.3
[25] cluster_2.1.8 R6_2.5.1 ica_1.0-3
[28] stringi_1.8.4 RColorBrewer_1.1-3 spatstat.data_3.1-4
[31] reticulate_1.40.0 parallelly_1.41.0 spatstat.univar_3.1-1
[34] lmtest_0.9-40 scattermore_1.2 tensor_1.5
[37] future.apply_1.11.3 zoo_1.8-12 sctransform_0.4.1
[40] httpuv_1.6.15 Matrix_1.6-5 splines_4.3.3
[43] igraph_2.1.2 tidyselect_1.2.1 abind_1.4-8
[46] spatstat.random_3.3-2 codetools_0.2-20 miniUI_0.1.1.1
[49] spatstat.explore_3.3-3 listenv_0.9.1 lattice_0.22-6
[52] tibble_3.2.1 plyr_1.8.9 shiny_1.10.0
[55] ROCR_1.0-11 Rtsne_0.17 future_1.34.0
[58] fastDummies_1.7.4 survival_3.8-3 polyclip_1.10-7
[61] fitdistrplus_1.2-1 pillar_1.10.0 KernSmooth_2.23-26
[64] plotly_4.10.4 generics_0.1.3 RcppHNSW_0.6.0
[67] ggplot2_3.5.1 munsell_0.5.1 scales_1.3.0
[70] globals_0.16.3 xtable_1.8-4 glue_1.8.0
[73] lazyeval_0.2.2 tools_4.3.3 data.table_1.16.4
[76] RSpectra_0.16-2 RANN_2.6.2 leiden_0.4.3.1
[79] dotCall64_1.2 cowplot_1.1.3 grid_4.3.3
[82] tidyr_1.3.1 colorspace_2.1-1 nlme_3.1-166
[85] patchwork_1.3.0 cli_3.6.3 spatstat.sparse_3.1-0
[88] spam_2.11-0 viridisLite_0.4.2 dplyr_1.1.4
[91] uwot_0.2.2 gtable_0.3.6 digest_0.6.37
[94] progressr_0.15.1 ggrepel_0.9.6 htmlwidgets_1.6.4
[97] farver_2.1.2 htmltools_0.5.8.1 lifecycle_1.0.4
[100] httr_1.4.7 mime_0.12 MASS_7.3-60