246 lines
7.2 KiB
R
246 lines
7.2 KiB
R
getwd()
|
|
setwd("~/git/LSHTM_analysis/scripts/plotting")
|
|
getwd()
|
|
|
|
#########################################################
|
|
# TASK: output barplot by position with each bar coloured by
|
|
# its stability value and active site positions indicated
|
|
# according to colour specified in "subcols_axis_PS.R"
|
|
#########################################################
|
|
|
|
#=======================================================================
|
|
|
|
############################################################
|
|
# 1: Installing and loading required packages and functions
|
|
############################################################
|
|
|
|
#source("Header_TT.R")
|
|
library(ggplot2)
|
|
library(data.table)
|
|
source("barplot_colour_function.R")
|
|
#source("subcols_axis.R")
|
|
source("subcols_axis_PS.R")
|
|
|
|
# should return the following dfs, directories and variables
|
|
# mut_pos_cols
|
|
# my_df
|
|
# my_df_u
|
|
# my_df_u_lig
|
|
# dup_muts
|
|
|
|
cat(paste0("Directories imported:"
|
|
, "\ndatadir:", datadir
|
|
, "\nindir:", indir
|
|
, "\noutdir:", outdir
|
|
, "\nplotdir:", plotdir))
|
|
|
|
cat(paste0("Variables imported:"
|
|
, "\ndrug:", drug
|
|
, "\ngene:", gene
|
|
, "\ngene_match:", gene_match
|
|
, "\nLength of upos:", length(upos)
|
|
, "\nAngstrom symbol:", angstroms_symbol))
|
|
|
|
# clear excess variable
|
|
rm(my_df, upos, dup_muts, my_df_u_lig)
|
|
#=======================================================================
|
|
#================
|
|
# Inspecting mut_pos_cols
|
|
# position numbers and colours
|
|
# open file from desktop ("sample_axis_cols") for cross checking
|
|
#================
|
|
|
|
table(mut_pos_cols$lab_bg)
|
|
check_lab_bg = sum( table(mut_pos_cols$lab_bg) ) == nrow(mut_pos_cols) # should be True
|
|
check_lab_bg
|
|
|
|
table(mut_pos_cols$lab_bg2)
|
|
check_lab_bg2 = sum( table(mut_pos_cols$lab_bg2) ) == nrow(mut_pos_cols) # should be True
|
|
check_lab_bg2
|
|
|
|
table(mut_pos_cols$lab_fg)
|
|
check_lab_fg = sum( table(mut_pos_cols$lab_fg) ) == nrow(mut_pos_cols) # should be True
|
|
check_lab_fg
|
|
|
|
# sanity checks:
|
|
if (check_lab_bg && check_lab_bg2 && check_lab_fg) {
|
|
print("PASS: No. of assigned colours match length")
|
|
}else{
|
|
print("FAIL: length of assigned colours mismatch")
|
|
quit()
|
|
}
|
|
|
|
# very important!
|
|
my_axis_colours = mut_pos_cols$lab_fg
|
|
|
|
# now clear mut_pos_cols
|
|
rm(mut_pos_cols)
|
|
#=======================================================================
|
|
#================
|
|
# Data for plots
|
|
#================
|
|
# REASSIGNMENT as necessary
|
|
df = my_df_u
|
|
|
|
# sanity checks
|
|
str(df)
|
|
|
|
###########################
|
|
# 2: Plot: DUET scores
|
|
###########################
|
|
|
|
#==========================
|
|
# Plot 2: Barplot with scores (unordered)
|
|
# corresponds to duet_outcome
|
|
# Stacked Barplot with colours: duet_outcome @ position coloured by
|
|
# stability scores. This is a barplot where each bar corresponds
|
|
# to a SNP and is coloured by its corresponding DUET stability value.
|
|
# Normalised values (range between -1 and 1 ) to aid visualisation
|
|
# NOTE: since barplot plots discrete values, colour = score, so number of
|
|
# colours will be equal to the no. of unique normalised scores
|
|
# rather than a continuous scale
|
|
# will require generating the colour scale separately.
|
|
#============================
|
|
# sanity checks
|
|
upos = unique(df$position)
|
|
|
|
table(df$duet_outcome)
|
|
table(df$duet_outcome)
|
|
|
|
# add frequency of positions (from lib data.table)
|
|
setDT(df)[, pos_count := .N, by = .(position)]
|
|
|
|
# this is cummulative
|
|
table(df$pos_count)
|
|
|
|
# use group by on this
|
|
library(dplyr)
|
|
snpsBYpos_df <- df %>%
|
|
group_by(position) %>%
|
|
summarize(snpsBYpos = mean(pos_count))
|
|
|
|
table(snpsBYpos_df$snpsBYpos)
|
|
|
|
snp_count = sort(unique(snpsBYpos_df$snpsBYpos))
|
|
|
|
# sanity checks
|
|
# should be a factor
|
|
df$duet_outcome = as.factor(df$duet_outcome)
|
|
is.factor(df$duet_outcome)
|
|
|
|
table(df$duet_outcome)
|
|
|
|
# should be -1 and 1
|
|
min(df$duet_scaled)
|
|
max(df$duet_scaled)
|
|
|
|
# sanity checks
|
|
# very important!!!!
|
|
tapply(df$duet_scaled, df$duet_outcome, min)
|
|
|
|
tapply(df$duet_scaled, df$duet_outcome, max)
|
|
|
|
# My colour FUNCTION: based on group and subgroup
|
|
# in my case;
|
|
# df = df
|
|
# group = duet_outcome
|
|
# subgroup = normalised score i.e duet_scaled
|
|
|
|
# check unique values in normalised data
|
|
u = unique(df$duet_scaled)
|
|
cat("No. of unique values in normalised data:", length(u))
|
|
|
|
#%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|
|
# Run this section if rounding is to be used
|
|
# specify number for rounding
|
|
#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$duet_outcome, "_", my_grp, sep = "")
|
|
|
|
# Call the function to create the palette based on the group defined above
|
|
colours <- ColourPalleteMulti(df, "duet_outcome", "my_grp")
|
|
print(paste0("Colour palette generated for: ", length(colours), " colours"))
|
|
my_title = "Protein stability (DUET)"
|
|
|
|
#========================
|
|
# plot with axis colours
|
|
#========================
|
|
class(df$lab_bg)
|
|
|
|
# define cartesian coord
|
|
my_xlim = length(unique(df$position)); my_xlim
|
|
|
|
# axis label size
|
|
my_xals = 18
|
|
my_yals = 18
|
|
|
|
# axes text size
|
|
my_xats = 14
|
|
my_yats = 18
|
|
|
|
#******************
|
|
# generate plot: with axis colours
|
|
#******************
|
|
# plot name and location
|
|
# outdir/ (should be imported from reading file)
|
|
plotdir = paste0(outdir, "/", "plots") #should be imported from reading file
|
|
print(paste0("plot will be in:", plotdir))
|
|
|
|
bp_aa_subcols_duet = "barplot_acoloured_PS.svg"
|
|
plot_bp_aa_subcols_duet = paste0(outdir, "/plots/", bp_aa_subcols_duet)
|
|
|
|
print(paste0("plot name:", plot_bp_aa_subcols_duet))
|
|
|
|
svg(plot_bp_aa_subcols_duet, width = 26, height = 4)
|
|
|
|
g = ggplot(df, aes(factor(position, ordered = T)))
|
|
|
|
outPlot = g +
|
|
coord_cartesian(xlim = c(1, my_xlim)
|
|
#, ylim = c(0, 6)
|
|
, ylim = c(0, max(snp_count))
|
|
, clip = "off") +
|
|
geom_bar(aes(fill = group), colour = "grey") +
|
|
scale_fill_manual(values = colours
|
|
, guide = "none") +
|
|
geom_tile(aes(,-0.8, width = 0.95, height = 0.85)
|
|
, fill = df$lab_bg) +
|
|
geom_tile(aes(,-1.2, width = 0.95, height = -0.2)
|
|
, fill = df$lab_bg2) +
|
|
|
|
# Here it"s important to specify that your axis goes from 1 to max number of levels
|
|
theme(axis.text.x = element_text(size = my_xats
|
|
, angle = 90
|
|
, hjust = 1
|
|
, vjust = 0.4
|
|
, colour = my_axis_colours)
|
|
, axis.text.y = element_text(size = my_yats
|
|
, angle = 0
|
|
, hjust = 1
|
|
, vjust = 0)
|
|
, axis.title.x = element_text(size = my_xals)
|
|
, axis.title.y = element_text(size = my_yals )
|
|
, axis.ticks.x = element_blank()) +
|
|
labs(title = ""
|
|
, x = "position"
|
|
, y = "Frequency")
|
|
|
|
print(outPlot)
|
|
dev.off()
|
|
|
|
#!!!!!!!!!!!!!!!!
|
|
#Warning message:
|
|
# Vectorized input to `element_text()` is not officially supported.
|
|
#Results may be unexpected or may change in future versions of ggplot2.
|
|
#!!!!!!!!!!!!!!!!!
|
|
|
|
# for sanity and good practice
|
|
#rm(df)
|