# count numbers for ML

source("~/git/LSHTM_analysis/config/alr.R")
#source("~/git/LSHTM_analysis/config/embb.R")
#source("~/git/LSHTM_analysis/config/gid.R")
#source("~/git/LSHTM_analysis/config/katg.R")
#source("~/git/LSHTM_analysis/config/pnca.R")
#source("~/git/LSHTM_analysis/config/rpob.R")

#############################
# GET the actual merged dfs
#############################
#source("~/git/LSHTM_analysis/scripts/plotting/get_plotting_dfs.R") 
source("~/git/LSHTM_analysis/scripts/plotting/get_ml_dfs.R") 

#############################
# Output files: merged data
#############################
outfile_merged_df3 = paste0(outdir, '/', tolower(gene), '_merged_df3.csv')
#outfile_merged_df2 = paste0(outdir, '/', tolower(gene), '_merged_df2.csv')

################################################
# Add acticve site indication
###############################################
merged_df2$active_site      = as.integer(merged_df2$position %in% active_aa_pos)
merged_df3$active_site      = as.integer(merged_df3$position %in% active_aa_pos)

# check 
cols_sel = c('mutationinformation', 'mutation_info_labels'
             #, 'dm_om_numeric'
             , 'dst', 'dst_mode')

check_mdf2 = merged_df2[, cols_sel]
check_mdf2T = table(check_mdf2$mutationinformation, check_mdf2$dst_mode)
ft_mdf2 = as.data.frame.matrix(check_mdf2T)

#==================
# CHECK: dst mode
#===================
dst_check = all((ft_mdf2[,1]==0)==(ft_mdf2[,2]!=0)); dst_check

#=======================
# CHECK: dst mode labels
#=======================
#table(merged_df2$mutation_info_labels_orig)
#table(merged_df2$mutation_info_labels_v1)
table(merged_df2$mutation_info_labels)

dst_check1 = table(merged_df2$dst_mode)[1] == table(merged_df2$mutation_info_labels)[2]
dst_check2 = table(merged_df2$dst_mode)[2] == table(merged_df2$mutation_info_labels)[1]

check12 = all(dst_check && all(dst_check1 == dst_check2))

if (check12) {
  cat('\nPASS: dst mode labels verified. merged_df3 CAN be trusted! ')
}else{
  stop('FAIL: Something is wrong with the dst_mode column. Quitting!')
}

table(is.na(merged_df3$dst))

#==========================
# CHECK: active site labels
#==========================
table(merged_df2$active_site)
table(merged_df3$active_site)
aa_check1 = all( table(merged_df2$active_site) == table(as.integer(merged_df2$position %in% active_aa_pos)) )
aa_check2 = all( table(merged_df3$active_site) == table(as.integer(merged_df3$position %in% active_aa_pos)) )             
                                                       
if ( all(aa_check1 && aa_check2) ){
  cat('\nActive site indications successfully applied to merged_dfs for gene:', tolower(gene))
}

gene
gene_match

nrow(merged_df3)

##############################################
write.csv(merged_df3, outfile_merged_df3)
#write.csv(merged_df2, outfile_merged_df2)
cat(paste("\nmerged df3 filename:", outfile_merged_df3
             #, "\nmerged df2 filename:", outfile_merged_df2)
             ))

#%%###################################################################

###################################################
###################################################
###################################################

# source("~/git/LSHTM_analysis/config/alr.R")
# source("~/git/LSHTM_analysis/config/embb.R")
# source("~/git/LSHTM_analysis/config/gid.R")
# source("~/git/LSHTM_analysis/config/katg.R")
# source("~/git/LSHTM_analysis/config/pnca.R")
# source("~/git/LSHTM_analysis/config/rpob.R")
# # 
df3_filename = paste0("~/git/Data/", drug, "/output/", tolower(gene), "_merged_df3.csv")
df3 = read.csv(df3_filename)
# # 
# mutationinformation
length(unique((df3$mutationinformation)))
# # 
# # #dm _om
# table(df3$mutation_info)
# #table(df3$mutation_info_orig)
# #table(df3$mutation_info_labels_orig)
# 
# # used in plots and analyses
# table(df3$mutation_info_labels) # different, and matches dst_mode
# table(df3$dst_mode)
#  
# # test_set
# na_count <-sapply(df3, function(y) sum(length(which(is.na(y)))))
# na_count[drug]
# # 
# # # training set
# table(df3[drug])
# # 
# # # drtype: MDR and XDR
# # #table(df3$drtype) orig i.e. incorrect ones!
# # table(df3$drtype_mode_labels)
# 
# 
# df3_complete = df3
# table(df3_complete$dst_mode)
# comp_lin_all = df3_complete[df3_complete$lineage_labels%in%c("L1", "L2", "L3", "L4"),]
# table(comp_lin_all$lineage); sum(table(comp_lin_all$lineage))
# 
# df3_actual =  df3[!is.na(df3$dst), ]
# table(df3_actual$dst_mode)
# comp_lin_actual = df3_actual[df3_actual$lineage_labels%in%c("L1", "L2", "L3", "L4"),]
# table(comp_lin_actual$lineage); sum(table(comp_lin_actual$lineage))
#