#!/usr/bin/env Rscript       
#source("~/git/LSHTM_analysis/config/alr.R")
source("~/git/LSHTM_analysis/config/embb.R")
#source("~/git/LSHTM_analysis/config/katg.R")
#source("~/git/LSHTM_analysis/config/gid.R")
#source("~/git/LSHTM_analysis/config/pnca.R")
#source("~/git/LSHTM_analysis/config/rpob.R")

# get plottting dfs 
source("~/git/LSHTM_analysis/scripts/plotting/get_plotting_dfs.R")
###################################################################
# FIXME: ADD distance to NA when SP replies
# DONE: plotting_globals.R
dist_columns = c("ligand_distance", "interface_dist")
DistCutOff = 10

common_cols  = c("mutationinformation"
                 , "X5uhc_position"
                 , "X5uhc_offset"
                 , "position"
                 , "dst_mode"
                 , "mutation_info_labels"
                 , "sensitivity", dist_columns )

#===================
# stability cols
#===================
raw_cols_stability =  c("duet_stability_change"
                        , "deepddg"
                        , "ddg_dynamut2"
                        , "ddg_foldx")

scaled_cols_stability = c("duet_scaled"       
                          , "deepddg_scaled"   
                          , "ddg_dynamut2_scaled"
                          , "foldx_scaled")

outcome_cols_stability = c("duet_outcome"
                           , "deepddg_outcome"
                           , "ddg_dynamut2_outcome"
                           , "foldx_outcome")

#===================
# affinity cols
#===================
raw_cols_affinity =  c("ligand_affinity_change"
                       , "mmcsm_lig"
                       , "mcsm_ppi2_affinity"
                       , "mcsm_na_affinity")

scaled_cols_affinity = c("affinity_scaled" 
                         , "mmcsm_lig_scaled" 
                         , "mcsm_ppi2_scaled" 
                         , "mcsm_na_scaled" )

outcome_cols_affinity  = c( "ligand_outcome"
                            , "mmcsm_lig_outcome"
                            , "mcsm_ppi2_outcome"
                            , "mcsm_na_outcome")
#===================
# conservation cols
#===================
raw_cols_conservation =  c("consurf_score"
                           , "snap2_score"
                           , "provean_score")

scaled_cols_conservation = c("consurf_scaled"
                             , "snap2_scaled"
                             , "provean_scaled")

# CANNOT strictly be used, as categories are not identical with conssurf missing altogether
outcome_cols_conservation = c("provean_outcome"
                              , "snap2_outcome"
                              , "consurf_colour_rev"
                              , "consurf_colour"#doesn't exist,use this mapping
)

all_cols = c(common_cols
            , raw_cols_stability
            , scaled_cols_stability
            , outcome_cols_stability
            , raw_cols_affinity
            , scaled_cols_affinity
            , outcome_cols_affinity
            , raw_cols_conservation
            , scaled_cols_conservation
            , outcome_cols_conservation)


#=======
# output
#=======
outdir_images = paste0("~/git/Writing/thesis/images/results/", tolower(gene), "/")

####################################
# merged_df3: NECESSARY pre-processing
###################################
df3 = merged_df3

#=================
# PREFORMATTING: for consistency
#=================
# DONE: combining_dfs.R
# df3$sensitivity = ifelse(df3$dst_mode == 1, "R", "S")
# table(df3$sensitivity)

# ConSurf labels
#consurf_colOld = "consurf_colour_rev"
#consurf_colNew = "consurf_outcome"
#df3[[consurf_colNew]] = df3[[consurf_colOld]]
#df3[[consurf_colNew]] = as.factor(df3[[consurf_colNew]])
#df3[[consurf_colNew]]
# not this bit
levels(df3$consurf_outcome) = c( "nsd", 1, 2, 3, 4, 5, 6, 7, 8, 9)
#levels(df3$consurf_outcome)

# SNAP2 labels
#snap2_colname = "snap2_outcome"
#df3[[snap2_colname]] <- str_replace(df3[[snap2_colname]], "effect", "Effect")
#df3[[snap2_colname]] <- str_replace(df3[[snap2_colname]], "neutral", "Neutral")

#  for ref: not needed perse as function already does this and assigns labels for barplots
# labels_duet = levels(as.factor(df3$duet_outcome))
# labels_foldx = levels(as.factor(df3$foldx_outcome))
# labels_deepddg = levels(as.factor(df3$deepddg_outcome))
# labels_ddg_dynamut2_outcome = levels(as.factor(df3$ddg_dynamut2_outcome))
# 
# labels_lig = levels(as.factor(df3_lig$ligand_outcome))
# labels_mmlig = levels(as.factor(df3_lig$mmcsm_lig_outcome))
# labels_ppi2 = levels(as.factor(df3_ppi2$mcsm_ppi2_outcome))
# 
# labels_provean = levels(as.factor(df3$provean_outcome))
# labels_snap2   = levels(as.factor(df3$snap2_outcome))
# labels_consurf = levels(as.factor(df3$consurf_colour_rev))
# df3$consurf_colour_rev = as.factor(df3$consurf_colour_rev )
##############################################################################
#######################################
# merged_df2: NECESSARY pre-processing
######################################
df2 = merged_df2

#=================
# PREFORMATTING: for consistency
#=================
# DONE: combining_dfs.R
# df2$sensitivity = ifelse(df2$dst_mode == 1, "R", "S")
# table(df2$sensitivity)

#----------------------------------------------------
# Create dst2: fill na in dst with value of dst_mode
# for epistasis
#----------------------------------------------------
# DONE: combining_dfs.R
# df2$dst2 = ifelse(is.na(df2$dst), df2$dst_mode, df2f$dst)

#----------------------------------------------------
# reverse signs for foldx scaled values for
# to allow average with other tools
#----------------------------------------------------
head(df2['ddg_foldx'])
df2['ddg_foldxC'] = abs(df2$ddg_foldx)
head(df2['ddg_foldxC'])

head(df2['foldx_scaled'])
df2['foldx_scaled_signC'] = abs(df2$foldx_scaled)
head(df2['foldx_scaled_signC'])

rm_foldx_cols = c("ddg_foldx","foldx_scaled")
raw_cols_stab_revised    = raw_cols_stability[!raw_cols_stability%in%rm_foldx_cols]
raw_cols_stab_revised    = c(raw_cols_stab_revised,"ddg_foldxC")

scaled_cols_stab_revised = scaled_cols_stability[!scaled_cols_stability%in%rm_foldx_cols]
scaled_cols_stab_revised = c(scaled_cols_stab_revised, "foldx_scaled_signC")

######################################################
# Affinity related variables
# DONE:in plotting_globals.R
# DistCutOff = 10
# LigDist_colname  # = "ligand_distance" # from globals 
# ppi2Dist_colname  = "interface_dist"
# naDist_colname    = "TBC"

######################################################
# corr colnames
# drug
# "dst_mode"
# "ligand_distance"
# "DUET"
# "mCSM-lig"       
# "FoldX"
# "DeepDDG"
# "ASA"
# "RSA"
# "KD"             
# "RD"
# "Consurf"
# "SNAP2"
# "MAF"             
# "Log (OR)"       
# "-Log (P)"
# "Dynamut2"
# "mCSM-PPI2"       
# "interface_dist" 

corr_ps_colnames = c("DUET"
, "FoldX"
, "DeepDDG"
, "Dynamut2"

, "MAF"             
, "Log (OR)"       
, "-Log (P)"

# , "ASA"
# , "RSA"
# , "KD"             
# , "RD"
# , "Consurf"
# , "SNAP2"

#, "mCSM-lig"       
#, "ligand_distance"
#, "mCSM-PPI2"       
#, "interface_dist" 
, "dst_mode"
, drug
)

corr_lig_colnames = c("mCSM-lig"       
                     , "MAF"             
                     , "Log (OR)"       
                     , "-Log (P)"
                     , "ligand_distance"
                     , "dst_mode"
                     , drug)

corr_ppi2_colnames = c("mCSM-PPI2"
                       , "SNAP2"             
                       , "Log (OR)"       
                       , "-Log (P)"
                       , "interface_dist" 
                       , "dst_mode"
                       , drug)

#FIXME: Add provean
corr_conservation_cols = c("Consurf"
                      , "SNAP2"
                       , "MAF"             
                       , "Log (OR)"       
                       , "-Log (P)"
                       , "dst_mode"
                       , drug)