Corrected-UMAP

1. load libraries

2. Load Seurat Object

#Load Seurat Object merged from cell lines and a control(PBMC) after filtration
load("AllSample_corrected.Robj")

AllSample

An object of class Seurat 
38620 features across 49193 samples within 3 assays 
Active assay: integrated (1991 features, 1991 variable features)
 2 layers present: data, scale.data
 2 other assays present: RNA, ADT
 3 dimensional reductions calculated: pca, umap, tsne

3. Perform PCA

ElbowPlot(AllSample)

NA
NA

4. Clustering

# understanding resolution
 AllSample <- FindClusters(AllSample, resolution = c(0.4, 0.5, 0.7, 1.2))

Modularity Optimizer version 1.3.0 by Ludo Waltman and Nees Jan van Eck

Number of nodes: 49193
Number of edges: 1512234

Running Louvain algorithm...

0%   10   20   30   40   50   60   70   80   90   100%
[----|----|----|----|----|----|----|----|----|----|
**************************************************|

Maximum modularity in 10 random starts: 0.8921
Number of communities: 18
Elapsed time: 15 seconds
Modularity Optimizer version 1.3.0 by Ludo Waltman and Nees Jan van Eck

Number of nodes: 49193
Number of edges: 1512234

Running Louvain algorithm...

0%   10   20   30   40   50   60   70   80   90   100%
[----|----|----|----|----|----|----|----|----|----|
**************************************************|

Maximum modularity in 10 random starts: 0.8838
Number of communities: 18
Elapsed time: 14 seconds
Modularity Optimizer version 1.3.0 by Ludo Waltman and Nees Jan van Eck

Number of nodes: 49193
Number of edges: 1512234

Running Louvain algorithm...

0%   10   20   30   40   50   60   70   80   90   100%
[----|----|----|----|----|----|----|----|----|----|
**************************************************|

Maximum modularity in 10 random starts: 0.8669
Number of communities: 19
Elapsed time: 12 seconds
Modularity Optimizer version 1.3.0 by Ludo Waltman and Nees Jan van Eck

Number of nodes: 49193
Number of edges: 1512234

Running Louvain algorithm...

0%   10   20   30   40   50   60   70   80   90   100%
[----|----|----|----|----|----|----|----|----|----|
**************************************************|

Maximum modularity in 10 random starts: 0.8375
Number of communities: 24
Elapsed time: 10 seconds

#5. Visualization

# non-linear dimensionality reduction --------------
# All_samples_Merged <- RunUMAP(All_samples_Merged, 
#                           dims = 1:8,
#                           verbose = FALSE)
#                                   

# note that you can set `label = TRUE` or use the LabelClusters function to help label
# individual clusters


DimPlot(AllSample,group.by = "orig.ident", 
        reduction = "umap",
        label.size = 3,
        repel = T,
        label = T, label.box = T)

DimPlot(AllSample,
        group.by = "integrated_snn_res.0.4", 
        reduction = "umap",
        label.size = 2,
        repel = T,
        label = T, label.box = T)

DimPlot(AllSample,
        group.by = "integrated_snn_res.0.5", 
        reduction = "umap",
        label.size = 2,
        repel = T,
        label = T, label.box = T)

DimPlot(AllSample,
        group.by = "integrated_snn_res.1.2", 
        reduction = "umap",
        label.size = 2,
        repel = T,
        label = T, label.box = T)

6. Cell Distribution Tables

cluster_table <- table(Idents(AllSample))
print(cluster_table)

   0    1    2    3    4    5    6    7    8    9   10   11   12   13   14   15   16   17   18   19   20   21   22   23 
5038 4655 4640 3724 3017 2960 2907 2839 2328 2178 1813 1742 1654 1525 1525 1073  900  891  858  825  817  493  444  347 

#0.4
cell_distribution_table <- table(AllSample$cell_line, AllSample$integrated_snn_res.0.4)
cell_distribution_df <- as.data.frame.matrix(cell_distribution_table)
print(cell_distribution_df)

#write.csv(cell_distribution_df, file = "/home/bioinfo/Downloads/integration_table_Marie-0.4.csv", row.names = TRUE)

#0.5
cell_distribution_table <- table(AllSample$cell_line, AllSample$integrated_snn_res.0.5)
cell_distribution_df <- as.data.frame.matrix(cell_distribution_table)
print(cell_distribution_df)

#write.csv(cell_distribution_df, file = "/home/bioinfo/Downloads/integration_table_Marie-0.4.csv", row.names = TRUE)

#0.7
cell_distribution_table <- table(AllSample$cell_line, AllSample$integrated_snn_res.0.7)
cell_distribution_df <- as.data.frame.matrix(cell_distribution_table)
print(cell_distribution_df)

#write.csv(cell_distribution_df, file = "/home/bioinfo/Downloads/integration_table_Marie-0.4.csv", row.names = TRUE)

#1.2
cell_distribution_table <- table(AllSample$cell_line, AllSample$integrated_snn_res.1.2)
cell_distribution_df <- as.data.frame.matrix(cell_distribution_table)
print(cell_distribution_df)

#write.csv(cell_distribution_df, file = "/home/bioinfo/Downloads/integration_table_Marie-0.4.csv", row.names = TRUE)

7. Find Top Genes (Markers)

Idents(AllSample) <- AllSample@meta.data$seurat_clusters

# find markers for every cluster compared to all remaining cells, report both positive and negative
SS.markers <- FindAllMarkers(AllSample,  
                              only.pos = FALSE,
                              logfc.threshold = 0.25, verbose = FALSE)
SS.markers

#write.table(SS.markers, file = "/home/bioinfo/Documents/1-SS-STeps/9-Integration/1-Marie_Uncorrected_Robj/SS_markers_all_clusters.tsv", sep = "\t", row.names = FALSE)

8. FeaturePlot

DefaultAssay(AllSample) <- "RNA"
myfeatures <- c("CD3", "CD3E", "CD4","CD8", "CD8A", "CD14", "CD19")
FeaturePlot(AllSample, reduction = "umap", dims = 1:2, features = myfeatures, ncol = 3, order = T) 

Warning: Could not find CD3 in the default search locations, found in 'ADT' assay insteadWarning: The following requested variables were not found: CD8

9. Save the Seurat object as an Robj file