DE with FC Scanner

1. load libraries

#Differential Expression Analysis

2. load seurat object

#Load Seurat Object L7
load("/home/bioinfo/0-imp_Robj/Harmony_integrated_All_samples_Merged_with_PBMC10x_with_harmony_clustering.Robj")
# sct_data <- GetAssayData(All_samples_Merged, assay = "SCT", layer = "data")

# memory.limit(size = 64000)
# 
# # Transpose the data so that cells are rows and genes are columns
# transposed_data <- t(as.data.frame(sct_data))
# 
# # Specify the file name and save as CSV
# write.csv(transposed_data, file = "table/SCT_data_All_samples_Merged_transposed.csv", row.names = TRUE)
# 
# 
# 
# 
# 
# # Extract metadata from Seurat object
# metadata <- All_samples_Merged@meta.data
# 
# # Write metadata to CSV
# write.csv(metadata, file = "Extra/Metadata_All_samples_Merged.csv", row.names = TRUE)

2. create PBMC CD4 T cells file

library(dplyr)

# Load your CSV file
# data <- read.csv("Extra/NewFiles/pbmc_METADATA.csv")
# 
# # Filter rows where all three predicted columns contain "CD4 T"
# filtered_data <- data %>%
#   filter(grepl("CD4 T", predicted.celltype.l1) & 
#          grepl("CD4 T", predicted.celltype.l2) & 
#          grepl("CD4 T", predicted.celltype.l3))
# 
# # Write the filtered data to a new CSV file, including the header
# write.csv(filtered_data, "CD4Tcells_PBMC_Control.csv", row.names = FALSE)
# 
# # Save the first column (PBMC cells) to a txt file without header
# write.table(filtered_data[, 1], "Extra/NewFiles/PBMC_CD4T_cells.txt", row.names = FALSE, col.names = FALSE, quote = FALSE)
# 
# # Load your CSV file
# data <- read.csv("Extra/NewFiles/Cell_lines/Metadata_All_cell_lines.csv")
# 
# 
# # Define the clusters of interest
# clusters_of_interest <- c(3, 8, 10, 18, 1, 2, 13, 4, 7, 9, 6, 16, 19)
# 
# # Filter cells based on the specified clusters
# filtered_data <- data %>%
#   filter(Harmony_snn_res.0.9 %in% clusters_of_interest)
# 
# # Write the filtered data to a new CSV file
# write.csv(filtered_data, "Extra/NewFiles/Cell_lines/filtered_cells_of_cell_lines_by_cluster.csv", row.names = FALSE)
# 
# # Save the first column (PBMC cells) without the header to a txt file
# write.table(filtered_data[, 1], "Extra/NewFiles/Cell_lines/All_cell_lines_cells.txt", row.names = FALSE, col.names = FALSE, quote = FALSE)
# 


# # Load your CSV file
# data <- read.csv("Extra/NewFiles/Cell_lines/Metadata_All_cell_lines.csv")
# 
# # To save P1 (L1+L2)
# # Define the clusters of interest
# clusters_of_interest <- c(3, 8, 10, 18)
# 
# # Filter cells based on the specified clusters
# filtered_data <- data %>%
#   filter(Harmony_snn_res.0.9 %in% clusters_of_interest)
# 
# # Write the filtered data to a new CSV file
# write.csv(filtered_data, "Extra/NewFiles/Patients_based_on_celllines(Clusters)/P1.csv", row.names = FALSE)
# 
# # Save the first column (PBMC cells) without the header to a txt file
# write.table(filtered_data[, 1], "Extra/NewFiles/Patients_based_on_celllines(Clusters)/P1.txt", row.names = FALSE, col.names = FALSE, quote = FALSE)
# 
# 
# # To save P1 (L3+L4)
# # Define the clusters of interest
# clusters_of_interest <- c(1, 2, 13)
# 
# # Filter cells based on the specified clusters
# filtered_data <- data %>%
#   filter(Harmony_snn_res.0.9 %in% clusters_of_interest)
# 
# # Write the filtered data to a new CSV file
# write.csv(filtered_data, "Extra/NewFiles/Patients_based_on_celllines(Clusters)/P2.csv", row.names = FALSE)
# 
# # Save the first column (PBMC cells) without the header to a txt file
# write.table(filtered_data[, 1], "Extra/NewFiles/Patients_based_on_celllines(Clusters)/P2.txt", row.names = FALSE, col.names = FALSE, quote = FALSE)
# 
# 
# 
# # To save P1 (L5+L6+L7)
# # Define the clusters of interest
# clusters_of_interest <- c(4, 7, 9, 6, 16, 19)
# 
# # Filter cells based on the specified clusters
# filtered_data <- data %>%
#   filter(Harmony_snn_res.0.9 %in% clusters_of_interest)
# 
# # Write the filtered data to a new CSV file
# write.csv(filtered_data, "Extra/NewFiles/Patients_based_on_celllines(Clusters)/P3.csv", row.names = FALSE)
# 
# # Save the first column (PBMC cells) without the header to a txt file
# write.table(filtered_data[, 1], "Extra/NewFiles/Patients_based_on_celllines(Clusters)/P3.txt", row.names = FALSE, col.names = FALSE, quote = FALSE)

#Differential Expression Analysis

3. FC SCanner for DE

library(foreach)
library(doParallel)

setwd("/isilon/homes/nabbasi/6-DE/17-SingleCellFCscanner/")

source("scFCscanner_028_in_list.r")

Script to calculate logFC, Filter logFC >2 & logfc <-1.5, exclude lines when mean_L1_L7 <0.2 & control_group <0.2

# #load libraries------------------------------------
# library(dplyr)
# library(tidyverse)
# 
# # Read the TSV file into R
# Exp_Allsample <- read_tsv("Extra/NewFiles/Results/Cell_lines_vs_CD4Tcells/SCT_All_cell_lines_cells_vs_PBMC_CD4T_cells.tsv")
# 
# # Calculate log-fold change using FC column in the file
# Exp_Allsample$log2FC <- log2(Exp_Allsample$FC_All_cell_lines_cells_PBMC_CD4T_cells  )
# 
# # Filter rows based on logFC criteria
# filtered_data <- Exp_Allsample %>% filter(log2FC > 3 | log2FC < -1)
# 
# # Exclude rows with "L1-L7 mean" and "mean control" both less than 0.2
# filtered_data_final <- filtered_data %>% filter(!(mean_All_cell_lines_cells < 0.2 & mean_PBMC_CD4T_cells < 0.2))
# 
# # Writing it to CSV file
# write.csv(filtered_data_final, "Extra/NewFiles/Results/Cell_lines_vs_CD4Tcells/filtered_data_Cell_lines_vs_CD4Tcells.csv", row.names = FALSE)

Script to calculate logFC, Filter logFC >2 & logfc <-1.5, exclude lines when mean_P1 <0.2 & control_group <0.2

# Load necessary libraries ------------------------------------
library(dplyr)
library(readr)  # 'readr' package is recommended for reading .csv and .tsv files efficiently

# Read the TSV file into R
Exp_Allsample <- read_tsv("Extra/NewFiles/Patients_based_on_celllines_Clusters/Results_3_comparisons/SCT_data_All_samples_Merged_transposed_tab_P2_vs_P3.csv")

# Check if the first column name is missing, and if so, set it to "gene"
if (colnames(Exp_Allsample)[1] == "...1") {
  colnames(Exp_Allsample)[1] <- "gene"
}

# Calculate log2 fold-change directly within the pipe (using mutate)
Exp_Allsample <- Exp_Allsample %>%
  mutate(log2FC = log2(FC_P2_P3))  # Using mutate to add log2FC as a new column

# Check the summary statistics and distribution of log2FC before filtering
summary(Exp_Allsample$log2FC)
hist(Exp_Allsample$log2FC, main="Distribution of log2FC", xlab="log2FC", col="lightblue", border="black")

# Choose your own log2FC threshold for both positive and negative values
threshold_positive <- 1  # Set your chosen threshold for positive log2FC
threshold_negative <- -1  # Set your chosen threshold for negative log2FC

# Filter rows based on log2FC criteria (separate thresholds for positive and negative log2FC)
filtered_data_final <- Exp_Allsample %>%
  filter(log2FC > threshold_positive | log2FC < threshold_negative)  # Filter based on both conditions


filtered_data_final <- filtered_data_final %>%
  filter(!(mean_P2 < 0.2 & mean_P3 < 0.2))  

# Write the filtered data to a CSV file
write.csv(filtered_data_final, "Extra/NewFiles/Patients_based_on_celllines_Clusters/Results_3_comparisons/2-filtered_P2_vs_P3.csv", row.names = FALSE)

To extract cell line according to clusters

# Load necessary libraries ------------------------------------
library(dplyr)
library(readr)  # 'readr' package is recommended for reading .csv and .tsv files efficiently

# Load your CSV file
data <- read.csv("/home/bioinfo/17-SingleCellFCscanner/Extra/NewFiles/Cell_lines/Metadata_All_cell_lines.csv")

# Define a helper function to filter and save data
filter_and_save <- function(data, clusters, cell_line, output_csv, output_txt) {
  # Filter data based on clusters and cell line
  filtered_data <- data %>%
    filter(Harmony_snn_res.0.9 %in% clusters, orig.ident == cell_line)

  # Save the filtered data as a CSV file
  write.csv(filtered_data, output_csv, row.names = FALSE)

  # Save the first column (e.g., cell identifiers) as a TXT file without a header
  write.table(filtered_data[, 1], output_txt, row.names = FALSE, col.names = FALSE, quote = FALSE)
}

# Filter for L1 in P1
filter_and_save(
  data = data,
  clusters = c(3, 8, 10, 18),  # Replace with clusters for L1
  cell_line = "L1",  # Specify the cell line (e.g., "L1")
  output_csv = "/home/bioinfo/17-SingleCellFCscanner/Extra/NewFiles/Cell_lines/P1_L1.csv",
  output_txt = "/home/bioinfo/17-SingleCellFCscanner/Extra/NewFiles/Cell_lines/P1_L1.txt"
)

# Filter for L2 in P1
filter_and_save(
  data = data,
  clusters = c(3, 8, 10, 18),  # Replace with clusters for L2
  cell_line = "L2",  # Specify the cell line (e.g., "L2")
  output_csv = "/home/bioinfo/17-SingleCellFCscanner/Extra/NewFiles/Cell_lines/P1_L2.csv",
  output_txt = "/home/bioinfo/17-SingleCellFCscanner/Extra/NewFiles/Cell_lines/P1_L2.txt"
)

# Filter for L3 in P2
filter_and_save(
  data = data,
  clusters = c(1, 2, 13),  # Replace with clusters for L3
  cell_line = "L3",  # Specify the cell line (e.g., "L3")
  output_csv = "/home/bioinfo/17-SingleCellFCscanner/Extra/NewFiles/Cell_lines/P2_L3.csv",
  output_txt = "/home/bioinfo/17-SingleCellFCscanner/Extra/NewFiles/Cell_lines/P2_L3.txt"
)

# Filter for L4 in P2
filter_and_save(
  data = data,
  clusters = c(1, 2, 13),  # Replace with clusters for L4
  cell_line = "L4",  # Specify the cell line (e.g., "L4")
  output_csv = "/home/bioinfo/17-SingleCellFCscanner/Extra/NewFiles/Cell_lines/P2_L4.csv",
  output_txt = "/home/bioinfo/17-SingleCellFCscanner/Extra/NewFiles/Cell_lines/P2_L4.txt"
)

# Filter for L5 in P3
filter_and_save(
  data = data,
  clusters = c(4, 7, 9, 6, 16, 19),  # Replace with clusters for L5
  cell_line = "L5",  # Specify the cell line (e.g., "L5")
  output_csv = "/home/bioinfo/17-SingleCellFCscanner/Extra/NewFiles/Cell_lines/P3_L5.csv",
  output_txt = "/home/bioinfo/17-SingleCellFCscanner/Extra/NewFiles/Cell_lines/P3_L5.txt"
)

# Filter for L6 in P3
filter_and_save(
  data = data,
  clusters = c(4, 7, 9, 6, 16, 19),  # Replace with clusters for L6
  cell_line = "L6",  # Specify the cell line (e.g., "L6")
  output_csv = "/home/bioinfo/17-SingleCellFCscanner/Extra/NewFiles/Cell_lines/P3_L6.csv",
  output_txt = "/home/bioinfo/17-SingleCellFCscanner/Extra/NewFiles/Cell_lines/P3_L6.txt"
)

# Filter for L7 in P3
filter_and_save(
  data = data,
  clusters = c(4, 7, 9, 6, 16, 19),  # Replace with clusters for L7
  cell_line = "L7",  # Specify the cell line (e.g., "L7")
  output_csv = "/home/bioinfo/17-SingleCellFCscanner/Extra/NewFiles/Cell_lines/P3_L7.csv",
  output_txt = "/home/bioinfo/17-SingleCellFCscanner/Extra/NewFiles/Cell_lines/P3_L7.txt"
)

FC SCanner for DE

library(foreach)
library(doParallel)

setwd("/home/bioinfo/17-SingleCellFCscanner/")

source("scFCscanner_028_in_list.r")

To extract cell line according to clusters

# Load necessary libraries ------------------------------------
library(dplyr)
library(readr)  # 'readr' package is recommended for reading .csv and .tsv files efficiently

# Read the TSV file into R
Exp_Allsample <- read_tsv("Extra/NewFiles/Cell_lines/cell_lines/cell_lines_vs_control/P2_L4_vs_PBMC_CD4T_cells.csv")

# Check if the first column name is missing, and if so, set it to "gene"
if (colnames(Exp_Allsample)[1] == "...1") {
  colnames(Exp_Allsample)[1] <- "gene"
}

# Calculate log2 fold-change directly within the pipe (using mutate)
Exp_Allsample <- Exp_Allsample %>%
  mutate(log2FC = log2(FC_P2_L4_PBMC_CD4T_cells))  # Using mutate to add log2FC as a new column

# Check the summary statistics and distribution of log2FC before filtering
summary(Exp_Allsample$log2FC)
hist(Exp_Allsample$log2FC, main="Distribution of log2FC", xlab="log2FC", col="lightblue", border="black")

# Choose your own log2FC threshold for both positive and negative values
threshold_positive <- 3.5  # Set your chosen threshold for positive log2FC
threshold_negative <- -1  # Set your chosen threshold for negative log2FC

# Filter rows based on log2FC criteria (separate thresholds for positive and negative log2FC)
filtered_data <- Exp_Allsample %>%
  filter(log2FC > threshold_positive | log2FC < threshold_negative)  # Filter based on both conditions


filtered_data_final <- filtered_data %>%
  filter(!(mean_P2_L4 < 0.2 & mean_PBMC_CD4T_cells < 0.2))  

# Write the filtered data to a CSV file
write.csv(filtered_data_final, "Extra/NewFiles/Cell_lines/cell_lines/cell_lines_vs_control/filtered_P2_L4_vs_PBMC_CD4T_cells.csv", row.names = FALSE)

To extract Clusters to compare to control

# Load necessary libraries ------------------------------------
library(dplyr)
library(readr)  # 'readr' package is recommended for reading .csv and .tsv files efficiently

# Load your CSV file
data <- read.csv("../17-SingleCellFCscanner/Extra/NewFiles/Cell_lines/Metadata_All_cell_lines.csv")


# # Define the clusters of interest
# clusters_of_interest <- c(3, 8, 10, 18, 1, 2, 13, 4, 7, 9, 6, 16, 19)

# Define the clusters of interest
clusters_of_interest <- c(19)

# Filter cells based on the specified clusters
filtered_data <- data %>%
  filter(Harmony_snn_res.0.9 %in% clusters_of_interest)

# Write the filtered data to a new CSV file
write.csv(filtered_data, "../17-SingleCellFCscanner/Extra/NewFiles/Clusters/filtered_cluster19.csv", row.names = FALSE)

# Save the first column (PBMC cells) without the header to a txt file
write.table(filtered_data[, 1], "Extra/NewFiles/Clusters/Cluster19_cells.txt", row.names = FALSE, col.names = FALSE, quote = FALSE)

To extract cell line according to clusters

# Load necessary libraries ------------------------------------
library(dplyr)
library(readr)  # 'readr' package is recommended for reading .csv and .tsv files efficiently

# Read the TSV file into R
Exp_Allsample <- read_tsv("Extra/NewFiles/Clusters/Clusters_vs_control/C19_cells_vs_PBMC_CD4T_cells.csv")

# Check if the first column name is missing, and if so, set it to "gene"
if (colnames(Exp_Allsample)[1] == "...1") {
  colnames(Exp_Allsample)[1] <- "gene"
}

# Calculate log2 fold-change directly within the pipe (using mutate)
Exp_Allsample <- Exp_Allsample %>%
  mutate(log2FC = log2(FC_Cluster19_cells_PBMC_CD4T_cells))  # Using mutate to add log2FC as a new column

# Check the summary statistics and distribution of log2FC before filtering
summary(Exp_Allsample$log2FC)
hist(Exp_Allsample$log2FC, main="Distribution of log2FC", xlab="log2FC", col="lightblue", border="black")

# Choose your own log2FC threshold for both positive and negative values
threshold_positive <- 4  # Set your chosen threshold for positive log2FC
threshold_negative <- -2  # Set your chosen threshold for negative log2FC

# Filter rows based on log2FC criteria (separate thresholds for positive and negative log2FC)
filtered_data <- Exp_Allsample %>%
  filter(log2FC > threshold_positive | log2FC < threshold_negative)  # Filter based on both conditions


filtered_data_final <- filtered_data %>%
  filter(!(mean_Cluster19_cells < 0.2 & mean_PBMC_CD4T_cells < 0.2))  

# Write the filtered data to a CSV file
write.csv(filtered_data_final, "Extra/NewFiles/Clusters/Clusters_vs_control/2-filtered_C19_vs_PBMC_CD4T_cells.csv", row.names = FALSE)

Barplot for up and down regulated genes

# Define file paths
input_file <- "Extra/NewFiles/Clusters/Clusters_vs_control/2-filtered_C19_vs_PBMC_CD4T_cells.csv"
output_folder <- "Extra/NewFiles/Clusters/Clusters_vs_control/Enrichment_files/"

# Ensure output folder exists
if (!dir.exists(output_folder)) {
  dir.create(output_folder, recursive = TRUE)
}

# Load the data (handling errors gracefully)
data <- tryCatch({
  read_csv(input_file)
}, error = function(e) {
  stop("Error reading input file. Check the file path or format.")
})

# Check if the necessary columns exist
if (!all(c("log2FC", "gene") %in% colnames(data))) {
  stop("Required columns ('log2FC' and 'gene') are missing in the data.")
}

# Filter for upregulated and downregulated genes
upregulated_genes <- data %>% filter(log2FC > 0)
downregulated_genes <- data %>% filter(log2FC < 0)

# Extract only gene names as vectors
upregulated_gene_names <- upregulated_genes %>% pull(gene)
downregulated_gene_names <- downregulated_genes %>% pull(gene)

# Define output file paths
upregulated_file <- file.path(output_folder, "upregulated_gene_names.txt")
downregulated_file <- file.path(output_folder, "downregulated_gene_names.txt")

# Save the gene names to text files
write_lines(upregulated_gene_names, upregulated_file)
write_lines(downregulated_gene_names, downregulated_file)

# Print the number of upregulated and downregulated genes
cat("Number of upregulated genes:", length(upregulated_gene_names), "\n")
cat("Number of downregulated genes:", length(downregulated_gene_names), "\n")
cat("Gene names saved to:\n")
cat("  Upregulated genes: ", upregulated_file, "\n")
cat("  Downregulated genes: ", downregulated_file, "\n")

create PBMC CD4 T cells file

library(dplyr)

# Load your CSV file
data <- read.csv("Extra/NewFiles/Cell_lines/filtered_cells_of_cell_lines_by_cluster.csv")

# Filter rows where all three predicted columns contain "CD4 T"
filtered_data <- data %>%
  filter(grepl("B memory", predicted.celltype.l2) &
         grepl("B memory lambda", predicted.celltype.l3))

# Write the filtered data to a new CSV file, including the header
write.csv(filtered_data, "Extra/NewFiles/B_memory_in_cellline_clusters.csv", row.names = FALSE)

# Save the first column (PBMC cells) to a txt file without header
write.table(filtered_data[, 1], "Extra/NewFiles/B_memory_cells_in_cellline_clusters.txt", row.names = FALSE, col.names = FALSE, quote = FALSE)

# # Load your CSV file
# data <- read.csv("Extra/NewFiles/Cell_lines/Metadata_All_cell_lines.csv")
# 
# 
# # Define the clusters of interest
# clusters_of_interest <- c(3, 8, 10, 18, 1, 2, 13, 4, 7, 9, 6, 16, 19)
# 
# # Filter cells based on the specified clusters
# filtered_data <- data %>%
#   filter(Harmony_snn_res.0.9 %in% clusters_of_interest)
# 
# # Write the filtered data to a new CSV file
# write.csv(filtered_data, "Extra/NewFiles/Cell_lines/filtered_cells_of_cell_lines_by_cluster.csv", row.names = FALSE)
# 
# # Save the first column (PBMC cells) without the header to a txt file
# write.table(filtered_data[, 1], "Extra/NewFiles/Cell_lines/All_cell_lines_cells.txt", row.names = FALSE, col.names = FALSE, quote = FALSE)
# 


# # Load your CSV file
# data <- read.csv("Extra/NewFiles/Cell_lines/Metadata_All_cell_lines.csv")
# 
# # To save P1 (L1+L2)
# # Define the clusters of interest
# clusters_of_interest <- c(3, 8, 10, 18)
# 
# # Filter cells based on the specified clusters
# filtered_data <- data %>%
#   filter(Harmony_snn_res.0.9 %in% clusters_of_interest)
# 
# # Write the filtered data to a new CSV file
# write.csv(filtered_data, "Extra/NewFiles/Patients_based_on_celllines(Clusters)/P1.csv", row.names = FALSE)
# 
# # Save the first column (PBMC cells) without the header to a txt file
# write.table(filtered_data[, 1], "Extra/NewFiles/Patients_based_on_celllines(Clusters)/P1.txt", row.names = FALSE, col.names = FALSE, quote = FALSE)
# 
# 
# # To save P1 (L3+L4)
# # Define the clusters of interest
# clusters_of_interest <- c(1, 2, 13)
# 
# # Filter cells based on the specified clusters
# filtered_data <- data %>%
#   filter(Harmony_snn_res.0.9 %in% clusters_of_interest)
# 
# # Write the filtered data to a new CSV file
# write.csv(filtered_data, "Extra/NewFiles/Patients_based_on_celllines(Clusters)/P2.csv", row.names = FALSE)
# 
# # Save the first column (PBMC cells) without the header to a txt file
# write.table(filtered_data[, 1], "Extra/NewFiles/Patients_based_on_celllines(Clusters)/P2.txt", row.names = FALSE, col.names = FALSE, quote = FALSE)
# 
# 
# 
# # To save P1 (L5+L6+L7)
# # Define the clusters of interest
# clusters_of_interest <- c(4, 7, 9, 6, 16, 19)
# 
# # Filter cells based on the specified clusters
# filtered_data <- data %>%
#   filter(Harmony_snn_res.0.9 %in% clusters_of_interest)
# 
# # Write the filtered data to a new CSV file
# write.csv(filtered_data, "Extra/NewFiles/Patients_based_on_celllines(Clusters)/P3.csv", row.names = FALSE)
# 
# # Save the first column (PBMC cells) without the header to a txt file
# write.table(filtered_data[, 1], "Extra/NewFiles/Patients_based_on_celllines(Clusters)/P3.txt", row.names = FALSE, col.names = FALSE, quote = FALSE)

FC SCanner for DE

library(foreach)
library(doParallel)

setwd("/home/bioinfo/17-SingleCellFCscanner/")

source("scFCscanner_028_in_list.r")

To extract cell line according to clusters

# Load necessary libraries ------------------------------------
library(dplyr)
library(readr)  # 'readr' package is recommended for reading .csv and .tsv files efficiently

# Read the TSV file into R
Exp_Allsample <- read_tsv("Extra/NewFiles/Other_cells_in_celllines/2-FC_scanner_Results/SCT_data_All_samples_Merged_transposed_tab_B_memory_cells_in_cellline_clusters_vs_PBMC_CD4T_cells.csv")

# Check if the first column name is missing, and if so, set it to "gene"
if (colnames(Exp_Allsample)[1] == "...1") {
  colnames(Exp_Allsample)[1] <- "gene"
}

# Calculate log2 fold-change directly within the pipe (using mutate)
Exp_Allsample <- Exp_Allsample %>%
  mutate(log2FC = log2(FC_B_memory_cells_in_cellline_clusters_PBMC_CD4T_cells))  # Using mutate to add log2FC as a new column

# Check the summary statistics and distribution of log2FC before filtering
summary(Exp_Allsample$log2FC)
hist(Exp_Allsample$log2FC, main="Distribution of log2FC", xlab="log2FC", col="lightblue", border="black")

# Choose your own log2FC threshold for both positive and negative values
threshold_positive <- 2.5  # Set your chosen threshold for positive log2FC
threshold_negative <- -1.5  # Set your chosen threshold for negative log2FC

# Filter rows based on log2FC criteria (separate thresholds for positive and negative log2FC)
filtered_data <- Exp_Allsample %>%
  filter(log2FC > threshold_positive | log2FC < threshold_negative)  # Filter based on both conditions


filtered_data_final <- filtered_data %>%
  filter(!(mean_B_memory_cells_in_cellline_clusters < 0.2 & mean_PBMC_CD4T_cells < 0.2))  

# Write the filtered data to a CSV file
write.csv(filtered_data_final, "Extra/NewFiles/Other_cells_in_celllines/3-files_for_Enrichment/2-filtered_B_memory_cells_in_cellline_clusters_vs_PBMC_CD4T_cells.csv", row.names = FALSE)

Barplot for up and down regulated genes

# Define file paths
input_file <- "Extra/NewFiles/Other_cells_in_celllines/3-files_for_Enrichment/2-filtered_NK_Proliferating_cells_in_cellline_clusters_vs_PBMC_CD4T_cells.csv"
output_folder <- "Extra/NewFiles/Other_cells_in_celllines/3-files_for_Enrichment/Enrichment_files/"

# Ensure output folder exists
if (!dir.exists(output_folder)) {
  dir.create(output_folder, recursive = TRUE)
}

# Load the data (handling errors gracefully)
data <- tryCatch({
  read_csv(input_file)
}, error = function(e) {
  stop("Error reading input file. Check the file path or format.")
})

# Check if the necessary columns exist
if (!all(c("log2FC", "gene") %in% colnames(data))) {
  stop("Required columns ('log2FC' and 'gene') are missing in the data.")
}

# Filter for upregulated and downregulated genes
upregulated_genes <- data %>% filter(log2FC > 0)
downregulated_genes <- data %>% filter(log2FC < 0)

# Extract only gene names as vectors
upregulated_gene_names <- upregulated_genes %>% pull(gene)
downregulated_gene_names <- downregulated_genes %>% pull(gene)

# Define output file paths
upregulated_file <- file.path(output_folder, "upregulated_gene_names.txt")
downregulated_file <- file.path(output_folder, "downregulated_gene_names.txt")

# Save the gene names to text files
write_lines(upregulated_gene_names, upregulated_file)
write_lines(downregulated_gene_names, downregulated_file)

# Print the number of upregulated and downregulated genes
cat("Number of upregulated genes:", length(upregulated_gene_names), "\n")
cat("Number of downregulated genes:", length(downregulated_gene_names), "\n")
cat("Gene names saved to:\n")
cat("  Upregulated genes: ", upregulated_file, "\n")
cat("  Downregulated genes: ", downregulated_file, "\n")

To extract cell line according to clusters

# Load necessary libraries ------------------------------------
library(dplyr)

Attachement du package : ‘dplyr’

Les objets suivants sont masqués depuis ‘package:stats’:

    filter, lag

Les objets suivants sont masqués depuis ‘package:base’:

    intersect, setdiff, setequal, union

library(readr)  # 'readr' package is recommended for reading .csv and .tsv files efficiently

# Read the TSV file into R
Exp_Allsample <- read_tsv("Extra/NewFiles/Results/Cell_lines_vs_CD4Tcells/SCT_All_cell_lines_cells_vs_PBMC_CD4T_cells.tsv")

New names:Rows: 27417 Columns: 8── Column specification ──────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────
Delimiter: "\t"
chr (1): ...1
dbl (7): mean_All_cell_lines_cells, mean_PBMC_CD4T_cells, Relative_variance_All_cell_lines_cells, Relative_variance_PBMC_CD4T_cells, FC_All_cell_lines_cel...
ℹ Use `spec()` to retrieve the full column specification for this data.
ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.

# Check if the first column name is missing, and if so, set it to "gene"
if (colnames(Exp_Allsample)[1] == "...1") {
  colnames(Exp_Allsample)[1] <- "gene"
}

# Calculate log2 fold-change directly within the pipe (using mutate)
Exp_Allsample <- Exp_Allsample %>%
  mutate(log2FC = log2(FC_All_cell_lines_cells_PBMC_CD4T_cells))  # Using mutate to add log2FC as a new column