spRandom-seq/01_Preprocessing_spRandom-seq_data.R at main · WangycLab/spRandom-seq · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
source("./Helper_functions.R")

visium_coord <- read.table("./Data/visium-v4_coordinates.txt")
colnames(visium_coord) <- c("Barcode", "X", "Y")
visium_coord <- visium_coord %>%
  arrange(X,Y)

########## Mouse Brain - 10X chip #######
dataObj_brain <- readRDS("./Input/10X_ST_Mouse_Brain_add_res.rds")
dataObj_brain@meta.data <- dataObj_brain@meta.data %>%
  rownames_to_column(var = "Barcode") %>%
  inner_join(., visium_coord, by = "Barcode") %>%
  column_to_rownames(var = "Barcode")

ggplot(data = dataObj_brain@meta.data,
       mapping = aes(x = X, y = Y, color = seurat_clusters)) +
  geom_point(size = 2.5) +
  scale_color_manual(values = cbp2) +
  coord_flip() +
  xlim(0,max(dataObj_brain@meta.data$X)+2) +
  ylim(0,max(dataObj_brain@meta.data$Y)+2) +
  theme_bw()

brain_10X_selected_spots <- process_json("./HE/V43J19-400-A1_BRAIN_SELECTED_SPOTS.json")
dataObj_brain@meta.data <- dataObj_brain@meta.data %>%
  rownames_to_column(var = "Barcode") %>%
  left_join(., brain_10X_selected_spots,
            by = c("X" = "col", "Y" = "row")) %>%
  column_to_rownames(var = "Barcode")

dataObj_brain_filtered <- subset(dataObj_brain, subset = tissue == 1)

dataObj_brain_filtered <- CreateSeuratObject(counts = dataObj_brain_filtered@assays$RNA@counts) %>%
  SCTransform(., ncells = 3000, verbose = FALSE) %>%
  RunPCA(verbose = FALSE) %>%
  RunUMAP(dims = 1:30) %>%
  FindNeighbors(dims = 1:30, verbose = FALSE) %>%
  FindClusters(verbose = FALSE)

dataObj_brain_filtered@meta.data <- dataObj_brain_filtered@meta.data %>%
  rownames_to_column(var = "Barcode") %>%
  inner_join(., visium_coord, by = "Barcode") %>%
  column_to_rownames(var = "Barcode")

ggplot(data = dataObj_brain_filtered@meta.data,
       mapping = aes(x = X, y = Y, color = seurat_clusters)) +
  geom_point(size = 2.5) +
  scale_color_manual(values = cbp2) +
  coord_flip() +
  xlim(0,max(dataObj_brain@meta.data$X)+2) +
  ylim(0,max(dataObj_brain@meta.data$Y)+2) +
  theme_bw()

DimPlot(dataObj_brain_filtered)
saveRDS(dataObj_brain_filtered, "./Data/SPA_Mouse_brain.rds")

########## Mouse Heart - 10X chip #######
dataObj_heart <- readRDS("./Input/10X_ST_Mouse_Heart_res.rds")
dataObj_heart@meta.data <- dataObj_heart@meta.data %>%
  rownames_to_column(var = "Barcode") %>%
  inner_join(., visium_coord, by = "Barcode") %>%
  column_to_rownames(var = "Barcode")

ggplot(data = dataObj_heart@meta.data,
       mapping = aes(x = X, y = Y, color = seurat_clusters)) +
  geom_point(size = 2.5) +
  scale_color_manual(values = cbp2) +
  coord_flip() +
  xlim(0,max(dataObj_heart@meta.data$X)+2) +
  ylim(0,max(dataObj_heart@meta.data$Y)+2) +
  theme_bw()

heart_10X_selected_spots <- process_json("./HE/V43A13-384-D1_HEART_SELECTED_SPOTS.json")
dataObj_heart@meta.data <- dataObj_heart@meta.data %>%
  rownames_to_column(var = "Barcode") %>%
  left_join(., heart_10X_selected_spots,
            by = c("X" = "col", "Y" = "row")) %>%
  column_to_rownames(var = "Barcode")

dataObj_heart_filtered <- subset(dataObj_heart, subset = tissue == 1)

dataObj_heart_filtered <- CreateSeuratObject(counts = dataObj_heart_filtered@assays$RNA@counts) %>%
  SCTransform(., ncells = 3000, verbose = FALSE) %>%
  RunPCA(verbose = FALSE) %>%
  RunUMAP(dims = 1:30) %>%
  FindNeighbors(dims = 1:30, verbose = FALSE) %>%
  FindClusters(verbose = FALSE)

dataObj_heart_filtered@meta.data <- dataObj_heart_filtered@meta.data %>%
  rownames_to_column(var = "Barcode") %>%
  inner_join(., visium_coord, by = "Barcode") %>%
  column_to_rownames(var = "Barcode")

ggplot(data = dataObj_heart_filtered@meta.data,
       mapping = aes(x = X, y = Y, color = seurat_clusters)) +
  geom_point(size = 2.5) +
  scale_color_manual(values = cbp2) +
  coord_flip() +
  xlim(0,max(dataObj_heart@meta.data$X)+2) +
  ylim(0,max(dataObj_heart@meta.data$Y)+2) +
  theme_bw()

DimPlot(dataObj_heart_filtered)
saveRDS(dataObj_heart_filtered, "./Data/SPA_Mouse_heart.rds")


########## hFB48 - BRCA Tumor ###########
walk(list.files("./Data/SPA_hFB48_add_raw/", full.names = T), function(f) gzip(f))
SPA_hFB48_add_check <- CreateSeuratObject(
  counts = Read10X("./Data/SPA_hFB48_add_newPipe_raw/"), min.cells = 1
)
SPA_hFB48_add_check[['RNA']] <- as(object = SPA_hFB48_add_check[['RNA']], Class = "Assay")
SPA_hFB48_add_check <- run_DP_GAT(SPA_hFB48_add_check, size = 64, human = T)
SPA_hFB48_add_check[['RNA']] <- as(object = SPA_hFB48_add_check[['RNA']], Class = "Assay")
Display_clust_spatial(SPA_hFB48_add_check, siz = 64)

UT_FB48 <- readxl::read_xlsx("./Under_Tissue/SPA_hFB48_2.xlsx") %>%
  pivot_longer(., colnames(.)[-1], names_to = "X", values_to = "under_tissue") %>%
  mutate(Y = as.numeric(Y),
         X = as.numeric(X)) %>%
  dplyr::select(X, Y, under_tissue)

SPA_hFB48 <- SPA_hFB48_add_check
SPA_hFB48@meta.data <- SPA_hFB48@meta.data %>%
  rownames_to_column(var = "SPOT_ID") %>%
  left_join(., UT_FB48, by = c("X", "Y")) %>%
  column_to_rownames(var = "SPOT_ID")

SPA_hFB48@meta.data[is.na(SPA_hFB48@meta.data)] <- 0
SPA_hFB48 <- subset(SPA_hFB48, subset = under_tissue == "1")
Display_clust_spatial(SPA_hFB48, siz = 64)

SPA_hFB48[['RNA']] <- as(object = SPA_hFB48[['RNA']], Class = "Assay")
SPA_hFB48 <- run_DP_GAT(SPA_hFB48, size = 64)
SPA_hFB48[['RNA']] <- as(object = SPA_hFB48[['RNA']], Class = "Assay")
Display_clust_spatial(SPA_hFB48, siz = 64)

median(SPA_hFB48$nCount_RNA)
median(SPA_hFB48$nFeature_RNA)

mean(SPA_hFB48$nCount_RNA)
mean(SPA_hFB48$nFeature_RNA)

saveRDS(SPA_hFB48, "./Data/SPA_hFB48_res.rds")