133 lines
4.3 KiB
R
Executable file
133 lines
4.3 KiB
R
Executable file
#!/usr/bin/env Rscript
|
|
#########################################################
|
|
# TASK: formatting data that will be used for various plots
|
|
#########################################################
|
|
# load libraries and functions
|
|
library(data.table)
|
|
library(dplyr)
|
|
#========================================================
|
|
# plotting_data(): formatting data for plots
|
|
# input args:
|
|
## input csv file
|
|
## lig cut off dist, default = 10 Ang
|
|
# output: list of 4 dfs, that need to be decompressed
|
|
## my_df
|
|
## my_df_u
|
|
## my_df_u_lig
|
|
## dup_muts
|
|
#========================================================
|
|
plotting_data <- function(df
|
|
, lig_dist_colname = 'ligand_distance'
|
|
, lig_dist_cutoff = 10) {
|
|
my_df = data.frame()
|
|
my_df_u = data.frame()
|
|
my_df_u_lig = data.frame()
|
|
dup_muts = data.frame()
|
|
|
|
#===========================
|
|
# Read file: struct params
|
|
#===========================
|
|
#df = read.csv(infile_params, header = T)
|
|
|
|
cat("\nInput dimensions:", dim(df))
|
|
|
|
#==================================
|
|
# add foldx outcome category
|
|
# and foldx scaled values
|
|
|
|
# This will enable to always have these variables available
|
|
# when calling for plots
|
|
# included this now in combine_dfs.py!!!! finallyS
|
|
#==================================
|
|
|
|
#------------------------------
|
|
# # adding foldx scaled values
|
|
# # scale data b/w -1 and 1
|
|
# #------------------------------
|
|
# n = which(colnames(df) == "ddg"); n
|
|
#
|
|
# my_min = min(df[,n]); my_min
|
|
# my_max = max(df[,n]); my_max
|
|
#
|
|
# df$foldx_scaled = ifelse(df[,n] < 0
|
|
# , df[,n]/abs(my_min)
|
|
# , df[,n]/my_max)
|
|
# # sanity check
|
|
# my_min = min(df$foldx_scaled); my_min
|
|
# my_max = max(df$foldx_scaled); my_max
|
|
#
|
|
# if (my_min == -1 && my_max == 1){
|
|
# cat("\nPASS: foldx ddg successfully scaled b/w -1 and 1"
|
|
# , "\nProceeding with assigning foldx outcome category")
|
|
# }else{
|
|
# cat("\nFAIL: could not scale foldx ddg values"
|
|
# , "Aborting!\n")
|
|
# }
|
|
|
|
#------------------------------
|
|
# adding foldx outcome category
|
|
# ddg<0 = "Stabilising" (-ve)
|
|
#------------------------------
|
|
# c1 = table(df$ddg < 0)
|
|
# df$foldx_outcome = ifelse(df$ddg < 0, "Stabilising", "Destabilising")
|
|
# c2 = table(df$ddg < 0)
|
|
#
|
|
# if ( all(c1 == c2) ){
|
|
# cat("\nPASS: foldx outcome successfully created")
|
|
# }else{
|
|
# cat("\nFAIL: foldx outcome could not be created. Aborting!\n")
|
|
# exit()
|
|
# }
|
|
|
|
#------------------------------
|
|
# renaming foldx column from
|
|
# "ddg" --> "ddg_foldx"
|
|
#------------------------------
|
|
|
|
# # change name to foldx
|
|
# colnames(df)[n] <- "ddg_foldx"
|
|
|
|
#==================================
|
|
# extract unique mutation entries
|
|
#==================================
|
|
|
|
# check for duplicate mutations
|
|
if ( length(unique(df$mutationinformation)) != length(df$mutationinformation)){
|
|
cat(paste0("\nCAUTION:", " Duplicate mutations identified"
|
|
, "\nExtracting these...\n"))
|
|
#cat(my_df[duplicated(my_df$mutationinformation),])
|
|
dup_muts = df[duplicated(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\n"))
|
|
my_df_u = df[!duplicated(df$mutationinformation),]
|
|
}else{
|
|
cat(paste0("\nNo duplicate mutations detected\n"))
|
|
my_df_u = df
|
|
}
|
|
|
|
upos = unique(my_df_u$position)
|
|
cat("\nDim of clean df:"); cat(dim(my_df_u), "\n")
|
|
cat("\nNo. of unique mutational positions:"); cat(length(upos), "\n")
|
|
|
|
#===============================================
|
|
# extract mutations <10 Angstroms and symbol
|
|
#===============================================
|
|
table(my_df_u[[lig_dist_colname]] < lig_dist_cutoff)
|
|
|
|
my_df_u_lig = my_df_u[my_df_u[[lig_dist_colname]] < lig_dist_cutoff,]
|
|
|
|
cat(paste0("There are ", nrow(my_df_u_lig), " sites lying within 10\u212b of the ligand\n"))
|
|
|
|
# return list of DFs
|
|
my_df = df
|
|
#df_names = c("my_df", "my_df_u", "my_df_u_lig", "dup_muts")
|
|
all_df = list(my_df, my_df_u, my_df_u_lig, dup_muts)
|
|
#all_df = Map(setNames, all_df, df_names)
|
|
|
|
return(all_df)
|
|
}
|
|
########################################################################
|
|
# end of data extraction and cleaning for plots #
|
|
########################################################################
|