sorted combining_dfs.py with all other data files and tidied up get_plotting_dfs.R
This commit is contained in:
parent
4ba4ff602e
commit
5c8a9e8f00
3 changed files with 111 additions and 76 deletions
|
|
@ -169,25 +169,31 @@ i_join = 'inner'
|
|||
#=====================
|
||||
# some preprocessing
|
||||
#=====================
|
||||
#-------------
|
||||
|
||||
#===========
|
||||
# FoldX
|
||||
#-------------
|
||||
#===========
|
||||
foldx_df.shape
|
||||
#=======================
|
||||
|
||||
#----------------------
|
||||
# scale foldx values
|
||||
#=======================
|
||||
#----------------------
|
||||
# rename ddg column to ddg_foldx
|
||||
foldx_df['ddg']
|
||||
foldx_df = foldx_df.rename(columns = {'ddg':'ddg_foldx'})
|
||||
foldx_df['ddg_foldx']
|
||||
|
||||
# Rescale values in Foldx_change col b/w -1 and 1 so negative numbers
|
||||
# stay neg and pos numbers stay positive
|
||||
foldx_min = foldx_df['ddg'].min()
|
||||
foldx_max = foldx_df['ddg'].max()
|
||||
foldx_min = foldx_df['ddg_foldx'].min()
|
||||
foldx_max = foldx_df['ddg_foldx'].max()
|
||||
foldx_min
|
||||
foldx_max
|
||||
|
||||
foldx_scale = lambda x : x/abs(foldx_min) if x < 0 else (x/foldx_max if x >= 0 else 'failed')
|
||||
|
||||
foldx_df['foldx_scaled'] = foldx_df['ddg'].apply(foldx_scale)
|
||||
print('Raw foldx scores:\n', foldx_df['ddg']
|
||||
foldx_df['foldx_scaled'] = foldx_df['ddg_foldx'].apply(foldx_scale)
|
||||
print('Raw foldx scores:\n', foldx_df['ddg_foldx']
|
||||
, '\n---------------------------------------------------------------'
|
||||
, '\nScaled foldx scores:\n', foldx_df['foldx_scaled'])
|
||||
|
||||
|
|
@ -195,8 +201,8 @@ print('Raw foldx scores:\n', foldx_df['ddg']
|
|||
fsmi = foldx_df['foldx_scaled'].min()
|
||||
fsma = foldx_df['foldx_scaled'].max()
|
||||
|
||||
c = foldx_df[foldx_df['ddg']>=0].count()
|
||||
foldx_pos = c.get(key = 'ddg')
|
||||
c = foldx_df[foldx_df['ddg_foldx']>=0].count()
|
||||
foldx_pos = c.get(key = 'ddg_foldx')
|
||||
|
||||
c2 = foldx_df[foldx_df['foldx_scaled']>=0].count()
|
||||
foldx_pos2 = c2.get(key = 'foldx_scaled')
|
||||
|
|
@ -209,20 +215,30 @@ else:
|
|||
, '\nGot:', foldx_pos2
|
||||
, '\n======================================================')
|
||||
|
||||
# rename ddg column to ddg_foldx
|
||||
foldx_df['ddg']
|
||||
foldx_df = foldx_df.rename(columns = {'ddg':'ddg_foldx'})
|
||||
foldx_df['ddg_foldx']
|
||||
#-------------------------
|
||||
# foldx outcome category
|
||||
#--------------------------
|
||||
foldx_df['foldx_outcome'] = foldx_df['ddg_foldx'].apply(lambda x: 'Stabilising' if x >= 0 else 'Destabilising')
|
||||
foldx_df[foldx_df['ddg_foldx']>=0].count()
|
||||
foc = foldx_df['foldx_outcome'].value_counts()
|
||||
|
||||
#-------------
|
||||
if foc['Stabilising'] == foldx_pos and foc['Stabilising'] == foldx_pos2:
|
||||
print('\nPASS: Foldx outcome category created')
|
||||
else:
|
||||
print('\nFAIL: Foldx outcome category could NOT be created'
|
||||
, '\nExpected number:', foldx_pos
|
||||
, '\nGot:', foc[0]
|
||||
, '\n======================================================')
|
||||
sys.exit()
|
||||
|
||||
#=======================
|
||||
# Deepddg
|
||||
#-------------
|
||||
#=======================
|
||||
deepddg_df.shape
|
||||
|
||||
#=======================
|
||||
#-------------------------
|
||||
# scale Deepddg values
|
||||
#=======================
|
||||
|
||||
#-------------------------
|
||||
# Rescale values in deepddg_change col b/w -1 and 1 so negative numbers
|
||||
# stay neg and pos numbers stay positive
|
||||
deepddg_min = deepddg_df['deepddg'].min()
|
||||
|
|
@ -252,6 +268,23 @@ else:
|
|||
, '\nExpected number:', deepddg_pos
|
||||
, '\nGot:', deepddg_pos2
|
||||
, '\n======================================================')
|
||||
|
||||
#--------------------------
|
||||
# Deepddg outcome category
|
||||
#--------------------------
|
||||
deepddg_df['deepddg_outcome'] = deepddg_df['deepddg'].apply(lambda x: 'Stabilising' if x >= 0 else 'Destabilising')
|
||||
deepddg_df[deepddg_df['deepddg']>=0].count()
|
||||
doc = deepddg_df['deepddg_outcome'].value_counts()
|
||||
|
||||
if doc['Stabilising'] == deepddg_pos and doc['Stabilising'] == deepddg_pos2:
|
||||
print('\nPASS: Deepddg outcome category created')
|
||||
else:
|
||||
print('\nFAIL: Deepddg outcome category could NOT be created'
|
||||
, '\nExpected number:', deepddg_pos
|
||||
, '\nGot:', doc[0]
|
||||
, '\n======================================================')
|
||||
sys.exit()
|
||||
|
||||
#%%=============================================================================
|
||||
# Now merges begin
|
||||
#%%=============================================================================
|
||||
|
|
|
|||
|
|
@ -16,7 +16,9 @@ library(dplyr)
|
|||
## my_df_u_lig
|
||||
## dup_muts
|
||||
#========================================================
|
||||
plotting_data <- function(df, lig_dist_colname = 'ligand_distance', lig_dist_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()
|
||||
|
|
@ -38,51 +40,51 @@ cat("\nInput dimensions:", dim(df))
|
|||
#==================================
|
||||
|
||||
#------------------------------
|
||||
# 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$foldx_scaled); my_max
|
||||
|
||||
if (my_min == -1 && my_max == 1){
|
||||
cat("\nPASS: foldx ddg successfully scaled b/w -1 and 1"
|
||||
, "\nProceeding with assigning foldx outcome category")
|
||||
}else{
|
||||
cat("\nFAIL: could not scale foldx ddg values"
|
||||
, "Aborting!\n")
|
||||
}
|
||||
# # adding foldx scaled values
|
||||
# # scale data b/w -1 and 1
|
||||
# #------------------------------
|
||||
# n = which(colnames(df) == "ddg"); n
|
||||
#
|
||||
# my_min = min(df[,n]); my_min
|
||||
# my_max = max(df[,n]); my_max
|
||||
#
|
||||
# df$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()
|
||||
}
|
||||
# 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"
|
||||
# # change name to foldx
|
||||
# colnames(df)[n] <- "ddg_foldx"
|
||||
|
||||
#==================================
|
||||
# extract unique mutation entries
|
||||
|
|
|
|||
|
|
@ -97,33 +97,33 @@ merged_df3_comp = all_plot_dfs[[4]]
|
|||
# adding deepddg scaled values
|
||||
# scale data b/w -1 and 1
|
||||
#============================
|
||||
n = which(colnames(merged_df3) == "deepddg"); n
|
||||
|
||||
my_min = min(merged_df3[,n]); my_min
|
||||
my_max = max(merged_df3[,n]); my_max
|
||||
|
||||
merged_df3$deepddg_scaled = ifelse(merged_df3[,n] < 0
|
||||
, merged_df3[,n]/abs(my_min)
|
||||
, merged_df3[,n]/my_max)
|
||||
# sanity check
|
||||
my_min = min(merged_df3$deepddg_scaled); my_min
|
||||
my_max = max(merged_df3$deepddg_scaled); my_max
|
||||
|
||||
if (my_min == -1 && my_max == 1){
|
||||
cat("\nPASS: DeepDDG successfully scaled b/w -1 and 1"
|
||||
#, "\nProceeding with assigning deep outcome category")
|
||||
, "\n")
|
||||
}else{
|
||||
cat("\nFAIL: could not scale DeepDDG ddg values"
|
||||
, "Aborting!")
|
||||
}
|
||||
|
||||
# n = which(colnames(merged_df3) == "deepddg"); n
|
||||
#
|
||||
# my_min = min(merged_df3[,n]); my_min
|
||||
# my_max = max(merged_df3[,n]); my_max
|
||||
#
|
||||
# merged_df3$deepddg_scaled = ifelse(merged_df3[,n] < 0
|
||||
# , merged_df3[,n]/abs(my_min)
|
||||
# , merged_df3[,n]/my_max)
|
||||
# # sanity check
|
||||
# my_min = min(merged_df3$deepddg_scaled); my_min
|
||||
# my_max = max(merged_df3$deepddg_scaled); my_max
|
||||
#
|
||||
# if (my_min == -1 && my_max == 1){
|
||||
# cat("\nPASS: DeepDDG successfully scaled b/w -1 and 1"
|
||||
# #, "\nProceeding with assigning deep outcome category")
|
||||
# , "\n")
|
||||
# }else{
|
||||
# cat("\nFAIL: could not scale DeepDDG ddg values"
|
||||
# , "Aborting!")
|
||||
# }
|
||||
#
|
||||
|
||||
####################################################################
|
||||
# Data for combining other dfs
|
||||
####################################################################
|
||||
|
||||
source("other_dfs_data.R")
|
||||
#source("other_dfs_data.R")
|
||||
|
||||
####################################################################
|
||||
# Data for subcols barplot (~heatmap)
|
||||
|
|
|
|||
Loading…
Add table
Add a link
Reference in a new issue