230 lines
6.7 KiB
R
Executable file
230 lines
6.7 KiB
R
Executable file
#!/usr/bin/env Rscript
|
|
#########################################################
|
|
# TASK: producing barplots
|
|
# basic barplots with count of mutations
|
|
# basic barplots with frequency of count of mutations
|
|
|
|
# Depends on
|
|
## plotting_globals.R (previously dir.R)
|
|
## plotting_data.R
|
|
#########################################################
|
|
# working dir
|
|
getwd()
|
|
setwd("~/git/LSHTM_analysis/scripts/plotting")
|
|
getwd()
|
|
|
|
# load libraries
|
|
#source("Header_TT.R")
|
|
library(ggplot2)
|
|
library(data.table)
|
|
library(dplyr)
|
|
require("getopt", quietly = TRUE) # cmd parse arguments
|
|
|
|
# load functions
|
|
source("plotting_globals.R")
|
|
source("plotting_data.R")
|
|
source("functions/stability_count_bp.R")
|
|
#########################################################
|
|
# command line args
|
|
#********************
|
|
# !!!FUTURE TODO!!!
|
|
# Can pass additional params of output/plot dir by user.
|
|
# Not strictly required for my workflow since it is optimised
|
|
# to have a streamlined input/output flow without filename worries.
|
|
#********************
|
|
spec = matrix(c(
|
|
"drug" ,"d", 1, "character",
|
|
"gene" ,"g", 1, "character",
|
|
"data" ,"f", 2, "character"
|
|
), byrow = TRUE, ncol = 4)
|
|
|
|
opt = getopt(spec)
|
|
|
|
#FIXME: detect if script running from cmd, then set these
|
|
drug = opt$drug
|
|
gene = opt$gene
|
|
infile = opt$data
|
|
|
|
# hardcoding when not using cmd
|
|
#drug = "streptomycin"
|
|
#gene = "gid"
|
|
|
|
if(is.null(drug)|is.null(gene)) {
|
|
stop("Missing arguments: --drug and --gene must both be specified (case-sensitive)")
|
|
}
|
|
#########################################################
|
|
# call functions with relevant args
|
|
#------------------------------------------
|
|
# import_dirs()
|
|
# should return the follwoing variables:
|
|
# datadir
|
|
# indir
|
|
# outdir
|
|
# plotdir
|
|
# dr_muts_col
|
|
# other_muts_col
|
|
# resistance_col
|
|
#--------------------------------------------
|
|
import_dirs(drug, gene)
|
|
#---------------------------------------------
|
|
# plotting_data()
|
|
# should return the following dfs:
|
|
# my_df
|
|
# my_df_u
|
|
# my_df_u_lig
|
|
# dup_muts
|
|
#----------------------------------------------
|
|
#infile = "/home/tanu/git/Data/streptomycin/output/gid_comb_stab_struc_params.csv"
|
|
#infile = ""
|
|
|
|
#if (!exists("infile") && exists("gene")){
|
|
if (!is.character(infile) && exists("gene")){
|
|
#in_filename_params = paste0(tolower(gene), "_all_params.csv")
|
|
in_filename_params = paste0(tolower(gene), "_comb_stab_struc_params.csv") # part combined for gid
|
|
infile = paste0(outdir, "/", in_filename_params)
|
|
cat("\nInput file not specified, assuming filename: ", infile, "\n")
|
|
}
|
|
|
|
# Get the DFs out of plotting_data()
|
|
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]]
|
|
|
|
#########################################################
|
|
cat(paste0("Directories imported:"
|
|
, "\ndatadir:", datadir
|
|
, "\nindir:", indir
|
|
, "\noutdir:", outdir
|
|
, "\nplotdir:", plotdir))
|
|
|
|
cat(paste0("\nVariables imported:"
|
|
, "\ndrug:", drug
|
|
, "\ngene:", gene
|
|
, "\n))
|
|
#, "\ngene_match:", gene_match
|
|
#, "\nLength of upos:", length(upos)
|
|
#, "\nAngstrom symbol:", angstroms_symbol))
|
|
#=======================================================================
|
|
#=======
|
|
# output
|
|
#=======
|
|
# plot 1
|
|
basic_bp_duet = paste0(tolower(gene), "_basic_barplot_PS.svg")
|
|
plot_basic_bp_duet = paste0(plotdir,"/", basic_bp_duet)
|
|
|
|
# plot 2
|
|
pos_count_duet = paste0(tolower(gene), "_position_count_PS.svg")
|
|
plot_pos_count_duet = paste0(plotdir, "/", pos_count_duet)
|
|
|
|
#=======================================================================
|
|
#================
|
|
# Data for plots
|
|
#================
|
|
# REASSIGNMENT as necessary
|
|
df = my_df_u
|
|
|
|
# sanity checks
|
|
str(df)
|
|
#=======================================================================
|
|
#****************
|
|
# Plot 1: Count of stabilising and destabilsing muts
|
|
#****************
|
|
svg(plot_basic_bp_duet)
|
|
print(paste0("plot1 filename:", basic_bp_duet))
|
|
|
|
#--------------
|
|
# start plot 1: call function
|
|
#--------------
|
|
stability_count_bp(plotdf = my_df_u
|
|
, df_colname = "duet_outcome"
|
|
, leg_title = "DUET outcome")
|
|
dev.off()
|
|
|
|
table(my_df_u$duet_outcome)
|
|
#=======================================================================
|
|
#****************
|
|
# Plot 2: frequency of positions
|
|
#****************
|
|
df_ncols = ncol(df)
|
|
df_nrows = nrow(df)
|
|
|
|
cat(paste0("original df dimensions:"
|
|
, "\nNo. of rows:", df_nrows
|
|
, "\nNo. of cols:", df_ncols
|
|
, "\nNow adding column: frequency of mutational positions"))
|
|
|
|
#setDT(df)[, .(pos_count := .N), by = .(position)]
|
|
setDT(df)[, pos_count := .N, by = .(position)]
|
|
|
|
rm(df_nrows, df_ncols)
|
|
|
|
df_nrows = nrow(df)
|
|
df_ncols = ncol(df)
|
|
|
|
cat(paste0("revised df dimensions:"
|
|
, "\nNo. of rows:", df_nrows
|
|
, "\nNo. of cols:", df_ncols))
|
|
|
|
# this is cummulative
|
|
table(df$pos_count)
|
|
|
|
# use group by on this
|
|
snpsBYpos_df <- df %>%
|
|
group_by(position) %>%
|
|
summarize(snpsBYpos = mean(pos_count))
|
|
|
|
table(snpsBYpos_df$snpsBYpos)
|
|
|
|
#!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
|
|
# FIXME, get this mutation_info, perhaps useful!
|
|
foo = select(df, mutationinformation
|
|
, wild_pos
|
|
, wild_type
|
|
, mutant_type
|
|
#, mutation_info # comes from meta data, notused yet
|
|
, position
|
|
, pos_count)
|
|
|
|
#write.csv(foo, "/pos_count_freq.csv")
|
|
#!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
|
|
|
|
#--------------
|
|
# start plot 2
|
|
#--------------
|
|
svg(plot_pos_count_duet)
|
|
print(paste0("plot filename:", plot_pos_count_duet))
|
|
my_ats = 25 # axis text size
|
|
my_als = 22 # axis label size
|
|
|
|
# to make x axis display all positions
|
|
# not sure if to use with sort or directly
|
|
my_x = sort(unique(snpsBYpos_df$snpsBYpos))
|
|
|
|
g = ggplot(snpsBYpos_df, aes(x = snpsBYpos))
|
|
OutPlot_pos_count = g + geom_bar(aes (alpha = 0.5)
|
|
, show.legend = FALSE) +
|
|
scale_x_continuous(breaks = unique(snpsBYpos_df$snpsBYpos)) +
|
|
#scale_x_continuous(breaks = my_x) +
|
|
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 nsSNPs"
|
|
, y = "Number of Sites")
|
|
|
|
print(OutPlot_pos_count)
|
|
|
|
dev.off()
|
|
########################################################################
|
|
# end of PS barplots
|
|
########################################################################
|