#!/usr/bin/env Rscript             
library(ggplot2)
library(tidyverse)
library(data.table)

setwd("~/git/LSHTM_analysis/scripts/functions/")
getwd()
#############################################################
#===========================================
# load functions, data, dirs, hardocded vars
# that will be used in testing the functions
#===========================================
source("plotting_data.R")
infile = "/home/tanu/git/Data/streptomycin/output/"
pd_df = plotting_data(infile)
my_df       = pd_df[[1]]
my_df_u     = pd_df[[2]]
my_df_u_lig = pd_df[[3]]
dup_muts    = pd_df[[4]]

source("../plotting_globals.R")
drug = "streptomycin"
gene = "gid"

import_dirs(drug, gene)

#=====================
# functions to test 
#=====================
source("stability_count_bp.R")
source("position_count_bp.R")
#################################################################
##############################################
# read a sample file containing muts and prop 
###############################################
df<- read.csv(file.choose())

setDT(df)[, pos_count := .N, by = .(position)]
foo = data.frame(df$position, df$pos_count)

#snpsBYpos_df <- df %>%
#  group_by(position) %>%
#  summarize(snpsBYpos = mean(pos_count))

# subset df without duplicates for position
df2 = df[!duplicated(df$position)]
##################################################################
# ---------------------------------------
# barplot for nssnps, coloured by aa prop
# ---------------------------------------
pos_colname = "position"
aa_prop_colname = "mut_prop_water"
aa_prop_colours = c("black", "blue")
my_legname = "aa_prop: water"

# call function
aa_prop_bp(plotdf = df
           , position_colname = pos_colname
           , fill_colname = aa_prop_colname
           , fill_colours = aa_prop_cols
           , leg_name = my_legname)
           
#===============================================================