#!/usr/bin/env Rscript             
#########################################################
# TASK: formatting data that will be used for various plots

# useful links
#https://stackoverflow.com/questions/38851592/r-append-column-in-a-dataframe-with-frequency-count-based-on-two-columns
#########################################################
# working dir and loading libraries
getwd()
setwd("~/git/LSHTM_analysis/scripts/plotting")
getwd()

#source("Header_TT.R")
library(ggplot2)
library(data.table)
library(dplyr)

source("dirs.R")

require("getopt", quietly = TRUE) #cmd parse arguments

#========================================================
# command line args
#spec = matrix(c(
#  "drug"   , "d", 1, "character",
#  "gene"   , "g", 1, "character"
#), byrow = TRUE, ncol = 4)

#opt = getopt(spec)

#drug = opt$druggene = opt$gene

#if(is.null(drug)|is.null(gene)) {
#  stop("Missing arguments: --drug and --gene must both be specified (case-sensitive)")
#}
#========================================================
#======
# input
#======
#in_filename = "mcsm_complex1_normalised.csv"
in_filename_params = paste0(tolower(gene), "_all_params.csv") 
infile_params = paste0(outdir, "/", in_filename_params)
cat(paste0("Input file 1:", infile_params) )


cat('columns based on variables:\n'
      , drug
      , '\n'
      , dr_muts_col
      , '\n'
      , other_muts_col
      , "\n"
      , resistance_col
      , '\n===============================================================')

#%%===============================================================
###########################
# Read file: struct params
###########################
cat("Reading struct params including mcsm:", in_filename_params)
    
my_df = read.csv(infile_params, header = T)

cat("\nInput dimensions:", dim(my_df)) 


###########################
# extract unique mutation entries
###########################

# check for duplicate mutations
if ( length(unique(my_df$mutationinformation)) != length(my_df$mutationinformation)){
  cat(paste0("\nCAUTION:", " Duplicate mutations identified"
             , "\nExtracting these..."))
  dup_muts = my_df[duplicated(my_df$mutationinformation),]
  dup_muts_nu = length(unique(dup_muts$mutationinformation))
  cat(paste0("\nDim of duplicate mutation df:", nrow(dup_muts)
             , "\nNo. of unique duplicate mutations:", dup_muts_nu
             , "\n\nExtracting df with unique mutations only"))
  my_df_u = my_df[!duplicated(my_df$mutationinformation),]
}else{
  cat(paste0("\nNo duplicate mutations detected"))
  my_df_u = my_df
}

upos = unique(my_df_u$position)
cat("\nDim of clean df:"); cat(dim(my_df_u))
cat("\nNo. of unique mutational positions:"); cat(length(upos), "\n")
  

###########################
# extract mutations <10Angstroms
###########################
table(my_df_u$ligand_distance<10)

my_df_u_lig = my_df_u[my_df_u$ligand_distance <10,]
angstroms_symbol = "\u212b"
cat(paste0("There are ", nrow(my_df_u_lig), " sites lying within 10", angstroms_symbol, " of the ligand\n"))

delta_symbol = "\u0394"; delta_symbol

########################################################################
#               end of data extraction and cleaning for plots          #
########################################################################
# clear variables
rm(opt, spec)