library(Seurat)
library(dplyr)

Full_final <- readRDS("/DATA/31/molteni/vexas_mouse_3/results/Full/01-seurat/Full_final.rds")

# computation on cluster numbers
setwd("/DATA/31/molteni/vexas_mouse_3/results/DE_markers/")
Idents(Full_final)<-"RNA_snn_h.orig.ident_res.0.6"

clusters <- levels(Full_final@meta.data$RNA_snn_h.orig.ident_res.0.6)


for(i in 0:length(clusters)) {
  markers <- FindMarkers(Full_final, ident.1 = "A", group.by = "orig.ident", subset.ident = i)
  markers['cluster'] <- i
  filename=paste("markers_cluster_",i,".csv", sep="")
  write.csv(markers, file=filename)
}

df <- list.files(path='/DATA/31/molteni/vexas_mouse_3/results/DE_markers/') %>% 
  lapply(read.csv) %>% 
  bind_rows

df <- df[df$p_val_adj<0.05,]

write.csv(df, file="AvsC_sigMarkers.csv")
cells_by_sample <- table(Full_final@meta.data$RNA_snn_h.orig.ident_res.0.6, Full_final@meta.data$orig.ident)
write.csv(cells_by_sample, file="cells_bysample.csv")


# computation on custom annotation

Full_final <- readRDS("/DATA/31/molteni/vexas_mouse_3/results/Full/01-seurat/Full_final_annot.rds")

setwd("/DATA/31/molteni/vexas_mouse_3/results/DE_markers2/new/")

clusters <- unique (Full_final$celltypes_annot)

for(i in clusters) {
  markers <- FindMarkers(Full_final, ident.1 = "A", group.by = "orig.ident", subset.ident = i)
  markers['cluster'] <- i
  filename=paste("markers_cluster_",i,".csv", sep="")
  write.csv(markers, file=filename)
}


df <- list.files(path='/DATA/31/molteni/vexas_mouse_3/results/DE_markers2/new/') %>% 
  lapply(read.csv) %>% 
  bind_rows

df <- df[df$p_val_adj<0.05,]
setwd("/DATA/31/molteni/vexas_mouse_3/results/summary")
write.csv(df, file="AvsC_sigMarkers_custom.csv")

cc_distribution <-as.data.frame (table(Full_final$Phase, Full_final$orig.ident, Full_final$celltypes_annot))
colnames(cc_distribution) <- c("Phase","orig.ident","cluster","CellNumb")
write.csv(cc_distribution, file="cell_cycle.csv")


 
# computation on custom annotation after filtering cluster 19, res 1.2

Full_final <- readRDS("/DATA/31/molteni/vexas_mouse_3/results/Full/01-seurat/Full_final_annot_filt.rds")

setwd("/DATA/31/molteni/vexas_mouse_3/results/DE_markers/AvsC/")

clusters <- unique (Full_final$celltypes_annot)

for(i in clusters) {
  markers <- FindMarkers(Full_final, ident.1 = "A", group.by = "orig.ident", subset.ident = i)
  markers['cluster'] <- i
  filename=paste("markers_cluster_",i,".csv", sep="")
  write.csv(markers, file=filename)
}


df <- list.files(path='/DATA/31/molteni/vexas_mouse_3/results/DE_markers/AvsC/') %>% 
  lapply(read.csv) %>% 
  bind_rows

df <- df[df$p_val_adj<0.05,]
setwd("/DATA/31/molteni/vexas_mouse_3/results/summary")
write.csv(df, file="AvsC_sigMarkers_custom.csv")

cells_by_sample <- table(Full_final@meta.data$celltypes_annot, Full_final@meta.data$orig.ident)
write.csv(cells_by_sample, file="cells_bysample.csv")

cc_distribution <-as.data.frame (table(Full_final$Phase, Full_final$orig.ident, Full_final$celltypes_annot))
colnames(cc_distribution) <- c("Phase","orig.ident","cluster","CellNumb")
write.csv(cc_distribution, file="cell_cycle.csv")


#markers tra cluster specifici
## GMP2 VS GMP2
setwd("/DATA/31/molteni/vexas_mouse_3/results/DE_GMP2_GMP1")
Idents(Full_final) <- "celltypes_annot"
markers <- FindMarkers(Full_final, ident.1 = "GMP_2", ident.2 = "GMP_1")
write.csv(markers, file="GMP2vsGMP1_Full.csv")

Idents(Full_final) <- "orig.ident"
Full_final_A <- subset(Full_final, idents = "A")
Idents(Full_final_A) <- "celltypes_annot"
markers <- FindMarkers(Full_final_A, ident.1 = "GMP_2", ident.2 = "GMP_1")
write.csv(markers, file="GMP2vsGMP1_A.csv")

Idents(Full_final) <- "orig.ident"
Full_final_C <- subset(Full_final, idents ="C")
Idents(Full_final_C) <- "celltypes_annot"
markers <- FindMarkers(Full_final_C, ident.1 = "GMP_2", ident.2 = "GMP_1")
write.csv(markers, file="GMP2vsGMP1_C.csv")


## B Cell Prog 1 vs B Cell Prog 2
setwd("/DATA/31/molteni/vexas_mouse_3/results/DE_BP1_BP2")
Idents(Full_final) <- "celltypes_annot"
markers <- FindMarkers(Full_final, ident.1 = "B_cell_progenitors_1", ident.2 = "B_cell_progenitors_2")
write.csv(markers, file="BP1vsBP2_Full.csv")

Idents(Full_final_A) <- "celltypes_annot"
markers <- FindMarkers(Full_final_A, ident.1 = "B_cell_progenitors_1", ident.2 = "B_cell_progenitors_2")
write.csv(markers, file="BP1vsBP2_A.csv")

Idents(Full_final_C) <- "celltypes_annot"
markers <- FindMarkers(Full_final_C, ident.1 = "B_cell_progenitors_1", ident.2 = "B_cell_progenitors_2")
write.csv(markers, file="BP1vsBP2_C.csv")


setwd("/DATA/31/molteni/vexas_mouse_3/results/DE_markers/AvsC/")
df <- list.files(path='/DATA/31/molteni/vexas_mouse_3/results/DE_markers/AvsC/') %>% 
  lapply(read.csv) %>% 
  bind_rows

#df <- df[df$p_val_adj<0.05,]

top50 <- df %>%
  group_by(cluster) %>%
  dplyr::filter(avg_log2FC > 0) %>%
  slice_head(n = 50) %>%
  ungroup() -> top50

write.csv(top50, file="top50_by_cluster.csv")