driverR.Rmd

---
title: "Elena 2024: Keratinocyte single-cell data
"
output: html_document
---

```{r}
library(Seurat)
library(tidyverse)
library(ggplot2)
library(FactoMineR)
library(factoextra)

library("AnnotationDbi")
library("org.Hs.eg.db")
library("biomaRt")
```

```{r}
# setwd('/mnt/data/users-home/thomas.sauter/Claudia_2023')
setwd('../Elena_2024')

# combined.all <- readRDS("combined.all_annotated.rds")
sdata <- readRDS("K82_K86_integrated_FINAL.rds")
```

```{r}
table(sdata$new_names)
prop.table(table(sdata$new_names))
table(sdata$orig.ident, sdata$new_names)
table(sdata$seurat_clusters)
table(sdata$orig.ident)

# Visualization
p <- DimPlot(sdata, reduction = "umap", label = TRUE)
print(p)
p1 <- DimPlot(sdata, reduction = "umap", label = TRUE, group.by = "orig.ident",pt.size = 1.5)
print(p1)

dim(sdata$RNA@counts)
```

```{r}
# Separating the clusters into different variables
maxNrCells <- 1000

counts_matrix <- sdata[["RNA"]]$counts
print(dim(counts_matrix))
listNames <- unique(sdata$new_names)
listNames
cell_types_to_extract <- c("H","M","P","TD1","TD2","F1","F2")
# cell_types_to_extract <- c("H")
cell_types_to_extract
table(sdata$new_names)

# for (ii in 1:length(cell_types_to_extract)) {
#   ii
#   subset_counts_matrix <- counts_matrix[ ,sdata$new_names %in% cell_types_to_extract[ii]]
#   if (ncol(subset_counts_matrix) > maxNrCells) {
#     subset_counts_matrix <- subset_counts_matrix[ , 1:maxNrCells]
#   }
#   print(dim(subset_counts_matrix))
#   write.table(subset_counts_matrix, file = paste("./Cluster_",ii,".txt",sep = ""))
# }
```

```{r}
mart <- useMart("ensembl")
mart <- useDataset("hsapiens_gene_ensembl", mart = mart)
head(listAttributes(mart))

EnsemblGeneID <- rownames(counts_matrix)

geneTable <- getBM(attributes= c("ensembl_gene_id", 
                                 "external_gene_name", "entrezgene_id"), 
                   values= EnsemblGeneID, mart= mart) 
head(geneTable)
head(geneTable$entrezgene_id)

for (ii in 1:length(cell_types_to_extract)) {
  ii
  subset_counts_matrix <- counts_matrix[ ,sdata$new_names %in% cell_types_to_extract[ii]]
  if (ncol(subset_counts_matrix) > maxNrCells) {
    subset_counts_matrix <- subset_counts_matrix[ , 1:maxNrCells]
  }
  
  data <- subset_counts_matrix
  
  m <- match(rownames(data),geneTable$external_gene_name)
  GeneID <- geneTable$entrezgene_id[m]
  Var1 <- GeneID
  head(rownames(data))
  head(GeneID)
  
  data2 <- cbind(EnsemblGeneID, GeneID,Var1,as.matrix(data[, 1:min(ncol(data),maxNrCells)]))
  head(data2)
  print(dim(data2))
  
  write.table(data2, file = paste("./Cluster_",ii,".txt",sep = ""), row.names = F)
}
```