Differential Expression Analysis of Malignant CD4Tcells vs Control(Normal CD4 Tcells)-GSEA-after_filtering_on_MeanExpression

1. load libraries

2. Perform DE analysis using Malignant_CD4Tcells_vs_Normal_CD4Tcells genes

markers <- read.csv("../1-MAST_with_SCT_batch_patient_cellline_as_Covariate_with_meanExpression.csv", header = T)

3. Summarize Markers

summarize_markers <- function(markers) {
  num_pval0 <- sum(markers$p_val_adj == 0)
  num_pval1 <- sum(markers$p_val_adj == 1)
  num_significant <- sum(markers$p_val_adj < 0.05)
  num_upregulated <- sum(markers$avg_log2FC > 1)
  num_downregulated <- sum(markers$avg_log2FC < -1)
  
  cat("Number of genes with p_val_adj = 0:", num_pval0, "\n")
  cat("Number of genes with p_val_adj = 1:", num_pval1, "\n")
  cat("Number of significant genes (p_val_adj < 0.05):", num_significant, "\n")
  cat("Number of upregulated genes (avg_log2FC > 1):", num_upregulated, "\n")
  cat("Number of downregulated genes (avg_log2FC < 1):", num_downregulated, "\n")
}

cat("Markers1 Summary (MAST with Batch Correction):\n")

summarize_markers(markers)

4. EnhancedVolcano plot

library(dplyr)
library(EnhancedVolcano)

# Assuming you have a data frame named Malignant_CD4Tcells_vs_Normal_CD4Tcells
# Filter genes based on lowest p-values but include all genes
filtered_genes <- markers %>%
  arrange(p_val_adj, desc(abs(avg_log2FC)))

# Create the EnhancedVolcano plot with the filtered data
EnhancedVolcano(
  filtered_genes, 
  lab = ifelse(filtered_genes$p_val_adj <= 0.0000905 & abs(filtered_genes$avg_log2FC) >= 1.0, filtered_genes$gene, NA),
  x = "avg_log2FC", 
  y = "p_val_adj",
  title = "Malignant CD4 T cells(cell lines) vs normal CD4 T cells",
  pCutoff = 0.0000905,
  FCcutoff = 1.0,
  legendPosition = 'right', 
  labCol = 'black',
  labFace = 'bold',
  boxedLabels = FALSE,  # Set to FALSE to remove boxed labels
  pointSize = 3.0,
  labSize = 5.0,
  col = c('grey70', 'black', 'blue', 'red'),  # Customize point colors
  selectLab = filtered_genes$gene[filtered_genes$p_val_adj <= 0.05 & abs(filtered_genes$avg_log2FC) >= 1.0]  # Only label significant genes
)

Warning: One or more p-values is 0. Converting to 10^-1 * current lowest non-zero p-value...

5. Filter and Summarize Results

# Apply the expression filter first
markers_filtered <- markers %>%
  filter(!(mean_expr_group1 < 0.2 & mean_expr_group2 < 0.2))


write.csv(markers_filtered, "3-MAST_with_SCT_batch_patient_cellline_as_Covariate_with_Filtered_on_meanExpression.csv", row.names = FALSE)

5.2 Summary Function

summarize_markers <- function(markers) {
  num_pval0 <- sum(markers$p_val_adj == 0)
  num_pval1 <- sum(markers$p_val_adj == 1)
  num_significant <- sum(markers$p_val_adj < 0.05)
  num_upregulated <- sum(markers$avg_log2FC > 1)
  num_downregulated <- sum(markers$avg_log2FC < -1)
  
  cat("Number of genes with p_val_adj = 0:", num_pval0, "\n")
  cat("Number of genes with p_val_adj = 1:", num_pval1, "\n")
  cat("Number of significant genes (p_val_adj < 0.05):", num_significant, "\n")
  cat("Number of upregulated genes (avg_log2FC > 1):", num_upregulated, "\n")
  cat("Number of downregulated genes (avg_log2FC < -1):", num_downregulated, "\n")
}

cat("Markers1 Summary (markers_mast_SCT):\n")

Markers1 Summary (markers_mast_SCT):

summarize_markers(markers_filtered)

Number of genes with p_val_adj = 0: 74 
Number of genes with p_val_adj = 1: 2209 
Number of significant genes (p_val_adj < 0.05): 3042 
Number of upregulated genes (avg_log2FC > 1): 4514 
Number of downregulated genes (avg_log2FC < -1): 125 

5.3 EnhancedVolcano plot

library(dplyr)
library(EnhancedVolcano)

# Assuming you have a data frame named Malignant_CD4Tcells_vs_Normal_CD4Tcells
# Filter genes based on lowest p-values but include all genes
filtered_genes <- markers_filtered %>%
  arrange(p_val_adj, desc(abs(avg_log2FC)))

# Create the EnhancedVolcano plot with the filtered data
EnhancedVolcano(
  filtered_genes, 
  lab = ifelse(filtered_genes$p_val_adj <= 0.0000905 & abs(filtered_genes$avg_log2FC) >= 1.0, filtered_genes$gene, NA),
  x = "avg_log2FC", 
  y = "p_val_adj",
  title = "Malignant CD4 T cells(cell lines) vs normal CD4 T cells",
  pCutoff = 0.0000905,
  FCcutoff = 1.0,
  legendPosition = 'right', 
  labCol = 'black',
  labFace = 'bold',
  boxedLabels = FALSE,  # Set to FALSE to remove boxed labels
  pointSize = 3.0,
  labSize = 5.0,
  col = c('grey70', 'black', 'blue', 'red'),  # Customize point colors
  selectLab = filtered_genes$gene[filtered_genes$p_val_adj <= 0.05 & abs(filtered_genes$avg_log2FC) >= 1.0]  # Only label significant genes
)

Warning: One or more p-values is 0. Converting to 10^-1 * current lowest non-zero p-value...