getwd()
setwd("~/git/LSHTM_analysis/mcsm_analysis/pyrazinamide/scripts/plotting")
getwd()

########################################################################
# 				Installing and loading required packages 			   #
########################################################################

source("../Header_TT.R")

#require(data.table)
#require(dplyr)

########################################################################
#		 Read file: call script for combining df for lig		   	   #
########################################################################

source("../combining_two_df_lig.R") 

#---------------------- PAY ATTENTION
# the above changes the working dir
#[1] "git/LSHTM_analysis/mcsm_analysis/pyrazinamide/scripts"
#---------------------- PAY ATTENTION

#==========================
# This will return:

# df with NA:
# merged_df2
# merged_df3

# df without NA:
# merged_df2_comp
# merged_df3_comp
#===========================

###########################
# Data for Lig plots
# you need merged_df3
# or
# merged_df3_comp
# since these have unique SNPs
# I prefer to use the merged_df3
# because using the _comp dataset means
# we lose some muts and at this level, we should use
# as much info as available
###########################

# uncomment as necessary
#<<<<<<<<<<<<<<<<<<<<<<<<<
# REASSIGNMENT
my_df  = merged_df3
#my_df = merged_df3_comp
#<<<<<<<<<<<<<<<<<<<<<<<<<

# delete variables not required
rm(merged_df2, merged_df2_comp, merged_df3, merged_df3_comp)

# quick checks
colnames(my_df)
str(my_df)

# Ensure correct data type in columns to plot: need to be factor
# sanity check
is.factor(my_df$Lig_outcome)
my_df$Lig_outcome = as.factor(my_df$lig_outcome)
is.factor(my_df$Lig_outcome)
#[1] TRUE

#############################
# Extra sanity check:
# for mcsm_lig ONLY
# Dis_lig_Ang should be <10
#############################

if (max(my_df$Dis_lig_Ang) < 10){
  print ("Sanity check passed: lig data is <10Ang")
}else{
  print ("Error: data should be filtered to be within 10Ang")
}

########################################################################
#               end of data extraction and cleaning for plots          #
########################################################################

#===========================
# Plot: Basic barplots 
#===========================

#===================
# Data for plots
#===================

#<<<<<<<<<<<<<<<<<<<<<<<<<
# REASSIGNMENT 
df  = my_df 
#<<<<<<<<<<<<<<<<<<<<<<<<<
rm(my_df)

# sanity checks
str(df)

if (identical(df$Position, df$position)){
  print("Sanity check passed: Columns 'Position' and 'position' are identical")
} else{
  print("Error!: Check column names and info contained")
}

#****************
# generate plot: No of stabilising and destabilsing muts
#****************
# set output dir for plots
getwd()
setwd("~/git/Data/pyrazinamide/output/plots")
getwd()

svg('basic_barplots_LIG.svg')

my_ats = 25 # axis text size
my_als = 22 # axis label size

# uncomment as necessary for either directly outputting results or 
# printing on the screen
g = ggplot(df, aes(x = Lig_outcome))
#prinfFile = g + geom_bar(
  g + geom_bar(
  aes(fill = Lig_outcome)
  , show.legend = TRUE
) + geom_label(
  stat = "count"
  , aes(label = ..count..)
  , color = "black"
  , show.legend = FALSE
  , size = 10) + theme(
    axis.text.x = element_blank()
    , axis.title.x = element_blank()
    , axis.title.y = element_text(size=my_als)
    , axis.text.y = element_text(size = my_ats)
    , legend.position = c(0.73,0.8)
    , legend.text = element_text(size=my_als-2)
    , legend.title = element_text(size=my_als)
    , plot.title = element_blank()
  ) + labs(
    title = ""
    , y = "Number of SNPs"
    #, fill='Ligand Outcome'
  )  + scale_fill_discrete(name = "Ligand Outcome"
                           , labels = c("Destabilising", "Stabilising"))
print(prinfFile)
dev.off()

#****************
# generate plot: No of positions
#****************
#get freq count of positions so you can subset freq<1
#require(data.table)
setDT(df)[, pos_count := .N, by = .(Position)] #169, 36

head(df$pos_count)
table(df$pos_count)
# this is cummulative
#1  2  3  4  5  6 
#5 24 36 56 30 18 

# use group by on this
snpsBYpos_df <- df %>%
  group_by(Position) %>%
  summarize(snpsBYpos = mean(pos_count)) 

table(snpsBYpos_df$snpsBYpos)
#1  2  3  4  5  6 
#5 12 12 14  6  3
# this is what will get plotted

svg('position_count_LIG.svg')

my_ats = 25 # axis text size
my_als = 22 # axis label size

g = ggplot(snpsBYpos_df, aes(x = snpsBYpos))
prinfFile = g + geom_bar(
  #g + geom_bar(
  aes (alpha = 0.5)
  , show.legend = FALSE
) +
  geom_label(
    stat = "count", aes(label = ..count..)
    , color = "black"
    , size = 10
  ) +
  theme( 
    axis.text.x = element_text(
      size = my_ats
      , angle = 0
    )
    , axis.text.y = element_text(
      size = my_ats
      , angle = 0
      , hjust = 1
    )
    , axis.title.x = element_text(size = my_als)
    , axis.title.y = element_text(size = my_als)
    , plot.title = element_blank()
  ) +
  labs(
    x = "Number of SNPs"
    , y = "Number of Sites"
  )
print(prinfFile)
dev.off()
########################################################################
#               			end of Lig barplots         			   #
########################################################################