tanu/LSHTM_analysis
1
0
Fork 0
Code Issues Pull requests Releases Wiki Activity

changes made to combining_dfs_plotting.R

Browse source
This commit is contained in:
Tanushree Tunstall 2021-06-23 16:15:15 +01:00
parent c6d1260f74
commit 2aec79af31
9 changed files with 258 additions and 126 deletions
Download patch file Download diff file Expand all files Collapse all files

45
scripts/functions/plotting_data.R
View file

@ -16,20 +16,18 @@ library(dplyr)
## my_df_u_lig
## dup_muts
#========================================================
plotting_data <- function(infile_params, mcsm_lig_cutoff = 10) {
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()
cat(paste0("\nInput file to prepare for plotting:", infile_params, "\n") )
#===========================
# Read file: struct params
#===========================
my_df = read.csv(infile_params, header = T)
#df = read.csv(infile_params, header = T)
cat("\nInput dimensions:", dim(my_df))
cat("\nInput dimensions:", dim(df))
#==================================
# add foldx outcome category
@ -43,17 +41,17 @@ cat("\nInput dimensions:", dim(my_df))
# adding foldx scaled values
# scale data b/w -1 and 1
#------------------------------
n = which(colnames(my_df) == "ddg"); n
n = which(colnames(df) == "ddg"); n
my_min = min(my_df[,n]); my_min
my_max = max(my_df[,n]); my_max
my_min = min(df[,n]); my_min
my_max = max(df[,n]); my_max
my_df$foldx_scaled = ifelse(my_df[,n] < 0
, my_df[,n]/abs(my_min)
, my_df[,n]/my_max)
df$foldx_scaled = ifelse(df[,n] < 0
, df[,n]/abs(my_min)
, df[,n]/my_max)
# sanity check
my_min = min(my_df$foldx_scaled); my_min
my_max = max(my_df$foldx_scaled); my_max
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"
@ -67,9 +65,9 @@ if (my_min == -1 && my_max == 1){
# adding foldx outcome category
# ddg<0 = "Stabilising" (-ve)
#------------------------------
c1 = table(my_df$ddg < 0)
my_df$foldx_outcome = ifelse(my_df$ddg < 0, "Stabilising", "Destabilising")
c2 = table(my_df$ddg < 0)
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")
@ -83,19 +81,19 @@ if ( all(c1 == c2) ){
#==================================
# check for duplicate mutations
if ( length(unique(my_df$mutationinformation)) != length(my_df$mutationinformation)){
if ( length(unique(df$mutationinformation)) != length(df$mutationinformation)){
cat(paste0("\nCAUTION:", " Duplicate mutations identified"
, "\nExtracting these...\n"))
#cat(my_df[duplicated(my_df$mutationinformation),])
dup_muts = my_df[duplicated(my_df$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 = my_df[!duplicated(my_df$mutationinformation),]
my_df_u = df[!duplicated(df$mutationinformation),]
}else{
cat(paste0("\nNo duplicate mutations detected\n"))
my_df_u = my_df
my_df_u = df
}
upos = unique(my_df_u$position)
@ -105,15 +103,14 @@ cat("\nNo. of unique mutational positions:"); cat(length(upos), "\n")
#===============================================
# extract mutations <10 Angstroms and symbol
#===============================================
table(my_df_u$ligand_distance<mcsm_lig_cutoff)
table(my_df_u$ligand_distance<lig_dist_cutoff)
my_df_u_lig = my_df_u[my_df_u$ligand_distance <mcsm_lig_cutoff,]
my_df_u_lig = my_df_u[my_df_u$ligand_distance <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
#return(list(my_df, my_df_u, my_df_u_lig, dup_muts))
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)