206 lines
6.4 KiB
R
206 lines
6.4 KiB
R
getwd()
|
|
setwd("~/git/LSHTM_analysis/scripts/plotting")
|
|
getwd()
|
|
|
|
#########################################################
|
|
# TASK:
|
|
|
|
#########################################################
|
|
|
|
########################################################################
|
|
# Installing and loading required packages and functions #
|
|
########################################################################
|
|
|
|
source("Header_TT.R")
|
|
source("barplot_colour_function.R")
|
|
|
|
########################################################################
|
|
# Read file: call script for combining df for PS #
|
|
########################################################################
|
|
#?????????????
|
|
#
|
|
########################################################
|
|
#%% variable assignment: input and output paths & filenames
|
|
drug = "pyrazinamide"
|
|
gene = "pncA"
|
|
gene_match = paste0(gene,"_p.")
|
|
cat(gene_match)
|
|
|
|
#=============
|
|
# directories
|
|
#=============
|
|
datadir = paste0("~/git/Data")
|
|
indir = paste0(datadir, "/", drug, "/input")
|
|
outdir = paste0("~/git/Data", "/", drug, "/output")
|
|
|
|
#======
|
|
# input
|
|
#======
|
|
#in_filename = "mcsm_complex1_normalised.csv"
|
|
in_filename_params = paste0(tolower(gene), "_all_params.csv")
|
|
infile_params = paste0(outdir, "/", in_filename_params)
|
|
cat(paste0("Input file:", infile_params) )
|
|
|
|
#=======
|
|
# output
|
|
#=======
|
|
subcols_bp_duet = "barplot_subcols_DUET.svg"
|
|
outPlot_subcols_bp_duet = paste0(outdir, "/plots/", subcols_bp_duet)
|
|
|
|
#%%===============================================================
|
|
###########################
|
|
# Read file: struct params
|
|
###########################
|
|
cat("Reading struct params including mcsm:", in_filename_params)
|
|
|
|
my_df = read.csv(infile_params
|
|
#, stringsAsFactors = F
|
|
, header = T)
|
|
|
|
cat("Input dimensions:", dim(my_df))
|
|
|
|
# clear variables
|
|
rm(in_filename_params, infile_params)
|
|
|
|
# quick checks
|
|
colnames(my_df)
|
|
str(my_df)
|
|
|
|
# check for duplicate mutations
|
|
if ( length(unique(my_df$mutationinformation)) != length(my_df$mutationinformation)){
|
|
cat(paste0("CAUTION:", " Duplicate mutations identified"
|
|
, "\nExtracting these..."))
|
|
dup_muts = my_df[duplicated(my_df$mutationinformation),]
|
|
dup_muts_nu = length(unique(dup_muts$mutationinformation))
|
|
cat(paste0("\nDim of duplicate mutation df:", nrow(dup_muts)
|
|
, "\nNo. of unique duplicate mutations:", dup_muts_nu
|
|
, "\n\nExtracting df with unique mutations only"))
|
|
my_df_u = my_df[!duplicated(my_df$mutationinformation),]
|
|
}else{
|
|
cat(paste0("No duplicate mutations detected"))
|
|
my_df_u = my_df
|
|
}
|
|
|
|
upos = unique(my_df_u$position)
|
|
cat("Dim of clean df:"); cat(dim(my_df_u))
|
|
cat("\nNo. of unique mutational positions:"); cat(length(upos))
|
|
|
|
########################################################################
|
|
# end of data extraction and cleaning for plots #
|
|
########################################################################
|
|
#===================
|
|
# Data for plots
|
|
#===================
|
|
# REASSIGNMENT as necessary
|
|
df = my_df_u
|
|
|
|
rm(my_df)
|
|
|
|
# sanity checks
|
|
upos = unique(df$position)
|
|
|
|
# should be a factor
|
|
is.factor(my_df$duet_outcome)
|
|
#[1] TRUE
|
|
|
|
table(df$duet_outcome)
|
|
|
|
# should be -1 and 1
|
|
min(df$duet_scaled)
|
|
max(df$duet_scaled)
|
|
|
|
tapply(df$duet_scaled, df$duet_outcome, min)
|
|
tapply(df$duet_scaled, df$duet_outcome, max)
|
|
|
|
#******************
|
|
# generate plot
|
|
#******************
|
|
#==========================
|
|
# Barplot with scores (unordered)
|
|
# corresponds to duet_outcome
|
|
# Stacked Barplot with colours: duet_outcome @ position coloured by
|
|
# stability scores. This is a barplot where each bar corresponds
|
|
# to a SNP and is coloured by its corresponding DUET stability value.
|
|
# Normalised values (range between -1 and 1 ) to aid visualisation
|
|
# NOTE: since barplot plots discrete values, colour = score, so number of
|
|
# colours will be equal to the no. of unique normalised scores
|
|
# rather than a continuous scale
|
|
# will require generating the colour scale separately.
|
|
#============================
|
|
|
|
# My colour FUNCTION: based on group and subgroup
|
|
# in my case;
|
|
# df = df
|
|
# group = duet_outcome
|
|
# subgroup = normalised score i.e duet_scaled
|
|
|
|
# check unique values in normalised data
|
|
u = unique(df$duet_scaled)
|
|
|
|
#%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|
|
# Run this section if rounding is to be used
|
|
n = 3
|
|
df$duet_scaledR = round(df$duet_scaled, n)
|
|
ur = unique(df$duet_scaledR)
|
|
|
|
# create an extra column called group which contains the "gp name and score"
|
|
# so colours can be generated for each unique values in this column
|
|
|
|
#my_grp = df$duet_scaledR # rounding
|
|
my_grp = df$duet_scaled # no rounding
|
|
#%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|
|
|
|
df$group <- paste0(df$duet_outcome, "_", my_grp, sep = "")
|
|
|
|
# Call the function to create the palette based on the group defined above
|
|
colours <- ColourPalleteMulti(df, "duet_outcome", "my_grp")
|
|
print(paste0("Colour palette generated for: ", length(colours), " colours"))
|
|
my_title = "Protein stability (DUET)"
|
|
|
|
# axis label size
|
|
my_xaxls = 13
|
|
my_yaxls = 15
|
|
|
|
# axes text size
|
|
my_xaxts = 15
|
|
my_yaxts = 15
|
|
|
|
#******************
|
|
# generate plot: NO axis colours
|
|
# no ordering of x-axis
|
|
#******************
|
|
# plot name and location
|
|
print(paste0("plot will be in:", outdir))
|
|
bp_subcols_duet = "barplot_coloured_PS.svg"
|
|
plot_bp_subcols_duet = paste0(outdir, "/plots/", bp_subcols_duet)
|
|
print(paste0("plot name:", plot_bp_subcols_duet))
|
|
|
|
svg(plot_bp_subcols_duet, width = 26, height = 4)
|
|
|
|
g = ggplot(df, aes(factor(position, ordered = T)))
|
|
outPlot = g +
|
|
geom_bar(aes(fill = group), colour = "grey") +
|
|
scale_fill_manual( values = colours
|
|
, guide = "none") +
|
|
theme( axis.text.x = element_text(size = my_xaxls
|
|
, angle = 90
|
|
, hjust = 1
|
|
, vjust = 0.4)
|
|
, axis.text.y = element_text(size = my_yaxls
|
|
, angle = 0
|
|
, hjust = 1
|
|
, vjust = 0)
|
|
, axis.title.x = element_text(size = my_xaxts)
|
|
, axis.title.y = element_text(size = my_yaxts ) ) +
|
|
labs(title = my_title
|
|
, x = "position"
|
|
, y = "Frequency")
|
|
|
|
print(outPlot)
|
|
dev.off()
|
|
# for sanity and good practice
|
|
rm(df)
|
|
#======================= end of plot
|
|
# axis colours labels
|
|
# https://stackoverflow.com/questions/38862303/customize-ggplot2-axis-labels-with-different-colors
|
|
# https://stackoverflow.com/questions/56543485/plot-coloured-boxes-around-axis-label
|