use Annotated Robj including PBMC10x to remove ILC and NK-just one Cell
Nasir Mahmood Abbasi
2025-01-14
#In this script I will I will remove Non T cells from PBMC
1. load libraries
2. Load Seurat Object
#Load Seurat Object merged from cell lines and a control(PBMC) after filtration
load("0-imp_Robj/All_Samples_Merged_with_10x_Azitmuth_Annotated_SCT_HPC_without_harmony_integration_removed_nonCD4cells_from_control_and_Bcells_from_L4.robj")
All_samples_MergedAn object of class Seurat
62931 features across 49388 samples within 6 assays
Active assay: SCT (26179 features, 3000 variable features)
3 layers present: counts, data, scale.data
5 other assays present: RNA, ADT, prediction.score.celltype.l1, prediction.score.celltype.l2, prediction.score.celltype.l3
4 dimensional reductions calculated: integrated_dr, ref.umap, pca, umap3. Cell type Distribution to check clusters
# We can apply it later on R obj to get these tables to compare it to decide about resolution.
# Azimuth l1
janitor::tabyl(All_samples_Merged@meta.data, predicted.celltype.l1, cell_line) predicted.celltype.l1 L1 L2 L3 L4 L5 L6 L7 PBMC PBMC_10x
B 0 0 12 4 35 85 121 0 0
CD4 T 5771 3124 6351 5967 4914 4575 4634 5171 3505
CD8 T 13 25 0 1 4 5 3 0 0
DC 0 0 1 4 29 12 41 0 0
Mono 0 0 1 13 3 0 5 0 0
NK 38 2784 6 25 11 259 38 0 0
other 0 0 57 8 1025 208 487 0 0
other T 3 2 0 1 1 4 2 0 0# Azimuth l2
janitor::tabyl(All_samples_Merged@meta.data, predicted.celltype.l2, cell_line) predicted.celltype.l2 L1 L2 L3 L4 L5 L6 L7 PBMC PBMC_10x
B intermediate 0 0 2 1 2 2 0 0 0
B memory 0 0 11 1 38 82 120 0 0
CD14 Mono 0 0 1 14 5 0 6 0 0
CD4 CTL 0 0 0 0 0 0 0 12 1
CD4 Naive 0 0 0 7 0 0 0 523 1512
CD4 Proliferating 2461 2852 5452 5391 4732 4002 4115 0 6
CD4 TCM 3320 270 887 562 178 557 517 4576 1963
CD4 TEM 1 0 0 0 0 0 0 60 23
CD8 Proliferating 0 0 0 0 0 1 1 0 0
CD8 TCM 1 16 0 0 0 0 0 0 0
CD8 TEM 1 8 0 0 2 3 1 0 0
cDC1 0 0 0 0 2 6 0 0 0
cDC2 0 0 0 4 11 3 35 0 0
dnT 2 3 0 1 2 5 2 0 0
HSPC 0 0 60 7 1035 213 490 0 0
ILC 0 0 0 1 0 0 0 0 0
NK 0 0 0 1 0 0 0 0 0
NK Proliferating 38 2785 6 24 11 259 38 0 0
Treg 1 1 9 9 4 15 6 0 0# Azimuth l3
janitor::tabyl(All_samples_Merged@meta.data, predicted.celltype.l3, cell_line) predicted.celltype.l3 L1 L2 L3 L4 L5 L6 L7 PBMC PBMC_10x
B intermediate lambda 0 0 2 0 4 2 3 0 0
B memory kappa 0 0 7 0 7 22 55 0 0
B memory lambda 0 0 3 1 7 29 26 0 0
CD14 Mono 0 0 1 15 9 0 22 0 0
CD4 CTL 0 0 0 0 0 0 0 14 1
CD4 Naive 0 0 0 9 0 0 0 526 1524
CD4 Proliferating 2462 2852 5452 5391 4732 4003 4115 0 6
CD4 TCM_1 1932 6 6 35 0 7 3 4038 1447
CD4 TCM_2 652 250 870 516 165 536 482 265 144
CD4 TCM_3 436 12 4 6 13 7 26 265 359
CD4 TEM_1 1 0 0 0 0 0 0 4 7
CD4 TEM_2 0 0 0 0 0 0 0 15 16
CD4 TEM_3 0 0 0 0 0 0 0 44 1
CD8 Naive_2 0 0 0 0 1 1 0 0 0
CD8 Proliferating 0 0 0 0 0 1 1 0 0
CD8 TCM_1 0 8 0 0 0 0 0 0 0
CD8 TCM_2 298 6 0 0 0 0 0 0 0
CD8 TCM_3 0 1 0 0 0 0 0 0 0
CD8 TEM_2 0 0 0 0 0 1 0 0 0
CD8 TEM_6 1 10 0 0 1 1 1 0 0
cDC1 0 0 0 1 17 36 16 0 0
cDC2_1 0 0 0 1 1 0 4 0 0
cDC2_2 0 0 1 3 12 4 33 0 0
dnT_2 2 3 0 2 2 6 2 0 0
HSPC 0 0 62 7 1036 214 493 0 0
ILC 0 0 1 3 0 0 0 0 0
NK Proliferating 38 2785 6 24 11 259 38 0 0
Treg Memory 3 2 13 9 4 19 11 0 04. filter cells just keep CD4 T cells
# Set identity to cell_line
Idents(All_samples_Merged) <- "cell_line"
# Remove ILC and NK cells based on predicted.celltype.l2
filtered_seurat <- subset(All_samples_Merged, subset = predicted.celltype.l2 != "ILC" & predicted.celltype.l2 != "NK")
library(ggplot2)
library(RColorBrewer)
# Assuming you have 10 different cell lines, generating a color palette with 10 colors
cell_line_colors <- brewer.pal(10, "Set3")
# Assuming All_samples_Merged$cell_line is a factor or character vector containing cell line names
data <- as.data.frame(table(filtered_seurat$cell_line))
colnames(data) <- c("cell_line", "nUMI") # Change column name to nUMI
ncells <- ggplot(data, aes(x = cell_line, y = nUMI, fill = cell_line)) +
geom_col() +
theme_classic() +
geom_text(aes(label = nUMI),
position = position_dodge(width = 0.9),
vjust = -0.25) +
scale_fill_manual(values = cell_line_colors) +
theme(axis.text.x = element_text(angle = 45, hjust = 1),
plot.title = element_text(hjust = 0.5)) + # Adjust the title position
ggtitle("Filtered cells per sample") +
xlab("Cell lines") + # Adjust x-axis label
ylab("Frequency") # Adjust y-axis label
print(ncells)
NA
NA
NACell type Distribution to check clusters
# We can apply it later on R obj to get these tables to compare it to decide about resolution.
# Azimuth l1
janitor::tabyl(filtered_seurat@meta.data, predicted.celltype.l1, cell_line) predicted.celltype.l1 L1 L2 L3 L4 L5 L6 L7 PBMC PBMC_10x
B 0 0 12 4 35 85 121 0 0
CD4 T 5771 3124 6351 5967 4914 4575 4634 5171 3505
CD8 T 13 25 0 1 4 5 3 0 0
DC 0 0 1 4 29 12 41 0 0
Mono 0 0 1 13 3 0 5 0 0
NK 38 2784 6 24 11 259 38 0 0
other 0 0 57 7 1025 208 487 0 0
other T 3 2 0 1 1 4 2 0 0janitor::tabyl(filtered_seurat@meta.data, predicted.celltype.l2, cell_line) predicted.celltype.l2 L1 L2 L3 L4 L5 L6 L7 PBMC PBMC_10x
B intermediate 0 0 2 1 2 2 0 0 0
B memory 0 0 11 1 38 82 120 0 0
CD14 Mono 0 0 1 14 5 0 6 0 0
CD4 CTL 0 0 0 0 0 0 0 12 1
CD4 Naive 0 0 0 7 0 0 0 523 1512
CD4 Proliferating 2461 2852 5452 5391 4732 4002 4115 0 6
CD4 TCM 3320 270 887 562 178 557 517 4576 1963
CD4 TEM 1 0 0 0 0 0 0 60 23
CD8 Proliferating 0 0 0 0 0 1 1 0 0
CD8 TCM 1 16 0 0 0 0 0 0 0
CD8 TEM 1 8 0 0 2 3 1 0 0
cDC1 0 0 0 0 2 6 0 0 0
cDC2 0 0 0 4 11 3 35 0 0
dnT 2 3 0 1 2 5 2 0 0
HSPC 0 0 60 7 1035 213 490 0 0
NK Proliferating 38 2785 6 24 11 259 38 0 0
Treg 1 1 9 9 4 15 6 0 0janitor::tabyl(filtered_seurat@meta.data, predicted.celltype.l3, cell_line) predicted.celltype.l3 L1 L2 L3 L4 L5 L6 L7 PBMC PBMC_10x
B intermediate lambda 0 0 2 0 4 2 3 0 0
B memory kappa 0 0 7 0 7 22 55 0 0
B memory lambda 0 0 3 1 7 29 26 0 0
CD14 Mono 0 0 1 14 9 0 22 0 0
CD4 CTL 0 0 0 0 0 0 0 14 1
CD4 Naive 0 0 0 9 0 0 0 526 1524
CD4 Proliferating 2462 2852 5452 5391 4732 4003 4115 0 6
CD4 TCM_1 1932 6 6 35 0 7 3 4038 1447
CD4 TCM_2 652 250 870 516 165 536 482 265 144
CD4 TCM_3 436 12 4 6 13 7 26 265 359
CD4 TEM_1 1 0 0 0 0 0 0 4 7
CD4 TEM_2 0 0 0 0 0 0 0 15 16
CD4 TEM_3 0 0 0 0 0 0 0 44 1
CD8 Naive_2 0 0 0 0 1 1 0 0 0
CD8 Proliferating 0 0 0 0 0 1 1 0 0
CD8 TCM_1 0 8 0 0 0 0 0 0 0
CD8 TCM_2 298 6 0 0 0 0 0 0 0
CD8 TCM_3 0 1 0 0 0 0 0 0 0
CD8 TEM_2 0 0 0 0 0 1 0 0 0
CD8 TEM_6 1 10 0 0 1 1 1 0 0
cDC1 0 0 0 1 17 36 16 0 0
cDC2_1 0 0 0 1 1 0 4 0 0
cDC2_2 0 0 1 3 12 4 33 0 0
dnT_2 2 3 0 2 2 6 2 0 0
HSPC 0 0 62 7 1036 214 493 0 0
ILC 0 0 1 2 0 0 0 0 0
NK Proliferating 38 2785 6 24 11 259 38 0 0
Treg Memory 3 2 13 9 4 19 11 0 03.Save the Seurat object as an Robj file
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