distribution_analysis.R

library(ggplot2)

#############################
# Load Data
#############################

base_path <- './'
file_name <- 'datasets/gan_datasets/tham_lasso_dataset.csv'
all_data <- read.csv(paste(base_path, file_name, sep=''), header=TRUE)

gene_list_file_name <- 'human_mouse_gene_lists/lasso_gene_list.csv'
gene_list <- read.csv(paste(base_path, gene_list_file_name, sep=''), header=TRUE)



################################
# Functions
################################

# Function from https://stackoverflow.com/questions/22742737/function-to-save-ggplot
# Special Thanks to enricoferrero
# https://stackoverflow.com/users/1540663/enricoferrero
savePlot <- function(myPlot, file_name) {
  png(paste(file_name, '.png', sep=''), width=732, height=543)
  print(myPlot)
  dev.off()
}

################################
# Variables
################################

# Vector of fields to examine
# columns_to_exclude <-  c('test', 'class')
# fields <- colnames(all_data)[match(columns_to_exclude, colnames(all_data))]


for (field in gene_list$Symbol) {
  if (field %in% colnames(all_data)) {
    the_title <- paste('Expression Historgram of', field)
    x_lab <- paste(field, 'Expression Value')
    the_plot <- ggplot(all_data, aes(x=all_data[ , field])) +
      geom_histogram() +
      geom_vline(xintercept=mean(all_data[ , field]), color='red', size=2) +
      labs(title=the_title, x=x_lab)
    print(the_plot)
    savePlot(the_plot, field)

    # Pause until ready for next
    readline("Press Enter for next")
  }
}