586 lines
19 KiB
R
Executable file
586 lines
19 KiB
R
Executable file
#!/usr/bin/env Rscript
|
|
#########################################################
|
|
# TASK: Barplots for mCSM DUET, ligand affinity, and foldX
|
|
# basic barplots with count of mutations
|
|
# basic barplots with frequency of count of mutations
|
|
|
|
# , df_colname = ""
|
|
# , leg_title = ""
|
|
# , ats = 25 # axis text size
|
|
# , als = 22 # axis label size
|
|
# , lts = 20 # legend text size
|
|
# , ltis = 22 # label title size
|
|
# , geom_ls = 10 # geom_label size
|
|
# , yaxis_title = "Number of nsSNPs"
|
|
# , bp_plot_title = ""
|
|
# , label_categories = c("Destabilising", "Stabilising")
|
|
# , title_colour = "chocolate4"
|
|
# , subtitle_text = NULL
|
|
# , sts = 20
|
|
# , subtitle_colour = "pink"
|
|
# #, leg_position = c(0.73,0.8) # within plot area
|
|
# , leg_position = "top"
|
|
# , bar_fill_values = c("#F8766D", "#00BFC4")
|
|
#########################################################
|
|
#=============
|
|
# Data: Input
|
|
#==============
|
|
#source("~/git/LSHTM_analysis/config/alr.R")
|
|
source("~/git/LSHTM_analysis/config/embb.R")
|
|
#source("~/git/LSHTM_analysis/config/katg.R")
|
|
#source("~/git/LSHTM_analysis/config/gid.R")
|
|
#source("~/git/LSHTM_analysis/config/pnca.R")
|
|
#source("~/git/LSHTM_analysis/config/rpob.R")
|
|
|
|
source("~/git/LSHTM_analysis/scripts/plotting/get_plotting_dfs.R")
|
|
source("~/git/LSHTM_analysis/scripts/plotting/plotting_colnames.R")
|
|
|
|
class(merged_df3)
|
|
merged_df3 = as.data.frame(merged_df3)
|
|
|
|
class(df3)
|
|
head(merged_df3$pos_count)
|
|
|
|
nc_pc_CHANGE = which(colnames(merged_df3)== "pos_count")
|
|
colnames(merged_df3)[nc_pc_CHANGE] = "df2_pos_count_all"
|
|
head(merged_df3$pos_count)
|
|
head(merged_df3$df2_pos_count_all)
|
|
|
|
# DROP pos_count column
|
|
# merged_df3$pos_count <-NULL
|
|
merged_df3 = merged_df3[, !colnames(merged_df3)%in%c("pos_count")]
|
|
head(merged_df3$pos_count)
|
|
|
|
df3 = merged_df3[, colnames(merged_df3)%in%plotting_cols]
|
|
|
|
|
|
|
|
#=======
|
|
# output
|
|
#=======
|
|
outdir_images = paste0("~/git/Writing/thesis/images/results/", tolower(gene), "/")
|
|
cat("plots will output to:", outdir_images)
|
|
|
|
###########################################################
|
|
# ConSurf labels
|
|
|
|
|
|
#------------------------------
|
|
# plot default sizes
|
|
#------------------------------
|
|
sts = 22
|
|
subtitle_colour = "black"
|
|
geom_ls = 10
|
|
##############################################################
|
|
#------------------------------
|
|
# stability barplots:
|
|
outcome_stability_cols
|
|
# label_categories should be = levels(as.factor(plot_df[[df_colname]]))
|
|
#-------------------------
|
|
|
|
# duetP
|
|
duetP = stability_count_bp(plotdf = df3
|
|
, df_colname = "duet_outcome"
|
|
, leg_title = "mCSM-DUET"
|
|
#, label_categories = labels_duet
|
|
, yaxis_title = "Number of nsSNPs"
|
|
, leg_position = "none"
|
|
, subtitle_text = "mCSM-DUET"
|
|
, geom_ls = geom_ls
|
|
, bar_fill_values = c("#F8766D", "#00BFC4")
|
|
, sts = sts
|
|
, subtitle_colour= subtitle_colour)
|
|
|
|
# foldx
|
|
foldxP = stability_count_bp(plotdf = df3
|
|
, df_colname = "foldx_outcome"
|
|
#, leg_title = "FoldX"
|
|
#, label_categories = labels_foldx
|
|
, yaxis_title = ""
|
|
, leg_position = "none"
|
|
, subtitle_text = "FoldX"
|
|
, geom_ls = geom_ls
|
|
, bar_fill_values = c("#F8766D", "#00BFC4")
|
|
, sts = sts
|
|
, subtitle_colour= subtitle_colour)
|
|
|
|
|
|
# deepddg
|
|
deepddgP = stability_count_bp(plotdf = df3
|
|
, df_colname = "deepddg_outcome"
|
|
#, leg_title = "DeepDDG"
|
|
#, label_categories = labels_deepddg
|
|
, yaxis_title = "Number of nsSNPs"
|
|
, leg_position = "none"
|
|
, subtitle_text = "DeepDDG"
|
|
, geom_ls = geom_ls
|
|
, bar_fill_values = c("#F8766D", "#00BFC4")
|
|
, sts = sts
|
|
, subtitle_colour= subtitle_colour)
|
|
|
|
|
|
# deepddg
|
|
dynamut2P = stability_count_bp(plotdf = df3
|
|
, df_colname = "ddg_dynamut2_outcome"
|
|
#, leg_title = "Dynamut2"
|
|
#, label_categories = labels_ddg_dynamut2_outcome
|
|
, yaxis_title = ""
|
|
, leg_position = "none"
|
|
, subtitle_text = "Dynamut2"
|
|
, geom_ls = geom_ls
|
|
, bar_fill_values = c("#F8766D", "#00BFC4")
|
|
, sts = sts
|
|
, subtitle_colour= subtitle_colour)
|
|
|
|
dynamut2P
|
|
|
|
# # extract common legend
|
|
# common_legend = get_legend(duetP +
|
|
# guides(color = guide_legend(nrow = 1)) +
|
|
# theme(legend.position = "top"))
|
|
#
|
|
# #==========================
|
|
# #output: STABILITY PLOTS
|
|
# #===========================
|
|
# bp_stability_CLP = paste0(outdir_images
|
|
# , tolower(gene)
|
|
# ,"_bp_stability_CL.svg")
|
|
#
|
|
# svg(bp_stability_CLP, width = 15, height = 12)
|
|
# print(paste0("plot filename:", bp_stability_CLP))
|
|
#
|
|
# cowplot::plot_grid(
|
|
# common_legend,
|
|
# cowplot::plot_grid(duetP, foldxP
|
|
# , deepddgP, dynamut2P
|
|
# , nrow = 2
|
|
# , ncol = 2
|
|
# #, labels = c("(a)", "(b)", "(c)", "(d)")
|
|
# , labels = "AUTO"
|
|
# , label_size = 25)
|
|
# , ncol = 1
|
|
# , nrow = 2
|
|
# , rel_heights = c(0.4/10,9/10))
|
|
#
|
|
# dev.off()
|
|
###########################################################
|
|
#=========================
|
|
# Conservation outcome
|
|
# check this var:
|
|
outcome_conservation_cols
|
|
all(df3$consurf_colour_rev == df3$consurf_outcome)
|
|
#df3["consurf_outcome"] = as.factor(df3["consurf_outcome"])
|
|
levels(df3[["consurf_outcome"]])
|
|
|
|
#==========================
|
|
table(df3$consurf_outcome)
|
|
ggplot(df3, aes_string(x = "consurf_outcome")) +
|
|
geom_bar(aes(fill = eval(parse(text = "consurf_outcome")))
|
|
, show.legend = TRUE) +
|
|
scale_fill_manual(name = ""
|
|
, values = consurf_colours
|
|
#, labels = levels(df3[["snap2_outcome"]])
|
|
)
|
|
|
|
|
|
# consurf# had to turn label categories off for consurf
|
|
consurfP = stability_count_bp(plotdf = df3
|
|
, df_colname = "consurf_outcome"
|
|
#, leg_title = "ConSurf"
|
|
#, label_categories = labels_consurf
|
|
, yaxis_title = "Number of nsSNPs"
|
|
, leg_position = "top"
|
|
, subtitle_text = "ConSurf"
|
|
, geom_ls = 5
|
|
, bar_fill_values = consurf_colours # from globals
|
|
, sts = sts
|
|
, subtitle_colour= subtitle_colour)
|
|
|
|
consurfP
|
|
|
|
# provean
|
|
proveanP = stability_count_bp(plotdf = df3
|
|
, df_colname = "provean_outcome"
|
|
#, leg_title = "PROVEAN"
|
|
#, label_categories = labels_provean
|
|
, yaxis_title = ""
|
|
, leg_position = "top"
|
|
, subtitle_text = "PROVEAN"
|
|
, geom_ls = geom_ls
|
|
, bar_fill_values = c("#D01C8B", "#F1B6DA") # light pink and deep
|
|
, sts = sts
|
|
, subtitle_colour= subtitle_colour)
|
|
|
|
# snap2
|
|
snap2P = stability_count_bp(plotdf = df3
|
|
, df_colname = "snap2_outcome"
|
|
#, leg_title = "SNAP2"
|
|
#, label_categories = labels_snap2
|
|
, yaxis_title = ""
|
|
, leg_position = "top"
|
|
, subtitle_text = "SNAP2"
|
|
, geom_ls = geom_ls
|
|
, bar_fill_values = c("#D01C8B", "#F1B6DA") # light pink and deep
|
|
, sts = sts
|
|
, subtitle_colour= subtitle_colour)
|
|
|
|
|
|
#============================
|
|
# output: CONSERVATION PLOTS
|
|
#============================
|
|
# bp_conservation_CLP = paste0(outdir_images
|
|
# ,tolower(gene)
|
|
# ,"_bp_conservation_CL.svg" )
|
|
#
|
|
# print(paste0("plot filename:", bp_conservation_CLP))
|
|
# svg(bp_conservation_CLP, width = 15, height = 6.5)
|
|
#
|
|
# cowplot::plot_grid(proveanP, snap2P, consurfP
|
|
# , nrow = 1
|
|
# , ncol = 3
|
|
# #, labels = c("(a)", "(b)", "(c)", "(d)")
|
|
# , labels = "AUTO"
|
|
# , label_size = 25
|
|
# #, rel_heights = c(0.4/10,9/10))
|
|
# , rel_widths = c(0.9, 0.9, 1.1))
|
|
#
|
|
#
|
|
# dev.off()
|
|
|
|
###########################################################
|
|
#=========================
|
|
# Affinity outcome
|
|
# check this var: outcome_cols_affinity
|
|
# get from preformatting or put in globals
|
|
#==========================
|
|
DistCutOff
|
|
LigDist_colname # = "ligand_distance" # from globals
|
|
ppi2Dist_colname
|
|
naDist_colname
|
|
|
|
###########################################################
|
|
# get plotting data within the distance
|
|
df3_lig = df3[df3[[LigDist_colname]]<DistCutOff,]
|
|
df3_ppi2 = df3[df3[[ppi2Dist_colname]]<DistCutOff,]
|
|
df3_na = df3[df3[[naDist_colname]]<DistCutOff,]
|
|
common_bp_title = paste0("Sites <", DistCutOff, angstroms_symbol)
|
|
|
|
#------------------------------
|
|
# barplot for ligand affinity:
|
|
# <10 Ang of ligand
|
|
#------------------------------
|
|
mLigP = stability_count_bp(plotdf = df3_lig
|
|
, df_colname = "ligand_outcome"
|
|
#, leg_title = "mCSM-lig"
|
|
#, label_categories = labels_lig
|
|
, yaxis_title = "Number of nsSNPs"
|
|
, leg_position = "none"
|
|
, subtitle_text = "mCSM-lig"
|
|
, geom_ls = geom_ls
|
|
, bar_fill_values = c("#F8766D", "#00BFC4")
|
|
, sts = sts
|
|
, subtitle_colour= subtitle_colour
|
|
#, bp_plot_title = paste(common_bp_title, "ligand")
|
|
)
|
|
|
|
#------------------------------
|
|
# barplot for ligand affinity:
|
|
# <10 Ang of ligand
|
|
# mmCSM-lig: will be the same no. of sites but the effect will be different
|
|
#------------------------------
|
|
mmLigP = stability_count_bp(plotdf = df3_lig
|
|
, df_colname = "mmcsm_lig_outcome"
|
|
#, leg_title = "mmCSM-lig"
|
|
#, label_categories = labels_mmlig
|
|
, yaxis_title = ""
|
|
, leg_position = "none"
|
|
, subtitle_text = "mmCSM-lig"
|
|
, geom_ls = geom_ls
|
|
, bar_fill_values = c("#F8766D", "#00BFC4")
|
|
, sts = sts
|
|
, subtitle_colour= subtitle_colour
|
|
#, bp_plot_title = paste(common_bp_title, "ligand")
|
|
)
|
|
|
|
#------------------------------
|
|
# barplot for ppi2 affinity
|
|
# <10 Ang of interface
|
|
#------------------------------
|
|
ppi2P = stability_count_bp(plotdf = df3_ppi2
|
|
, df_colname = "mcsm_ppi2_outcome"
|
|
#, leg_title = "mCSM-ppi2"
|
|
#, label_categories = labels_ppi2
|
|
, yaxis_title = ""
|
|
, leg_position = "none"
|
|
, subtitle_text = "mCSM-ppi2"
|
|
, geom_ls = geom_ls
|
|
, bar_fill_values = c("#F8766D", "#00BFC4")
|
|
, sts = sts
|
|
, subtitle_colour= subtitle_colour
|
|
, bp_plot_title = paste(common_bp_title, "interface")
|
|
)
|
|
|
|
# # extract common legend
|
|
# common_legend_aff = get_legend(mLigP +
|
|
# guides(color = guide_legend(nrow = 1)) +
|
|
# theme(legend.position = "top"))
|
|
#
|
|
# #==========================
|
|
# # output: AFFINITY PLOTS
|
|
# #==========================
|
|
# bp_affinity_CLP = paste0(outdir_images
|
|
# ,tolower(gene)
|
|
# ,"_bp_affinity_CL.svg" )
|
|
#
|
|
# print(paste0("plot filename:", bp_stability_CLP))
|
|
# svg(bp_affinity_CLP, width = 15, height = 6.5)
|
|
#
|
|
# cowplot::plot_grid(
|
|
# common_legend,
|
|
# cowplot::plot_grid(mLigP, mmLigP
|
|
# , ppi2P
|
|
# , nrow = 1
|
|
# , ncol = 3
|
|
# #, labels = c("(a)", "(b)", "(c)", "(d)")
|
|
# , labels = "AUTO"
|
|
# , label_size = 25)
|
|
# , ncol = 1
|
|
# , nrow = 2
|
|
# , rel_heights = c(0.4/10,9/10))
|
|
# #, rel_widths = c(1,1,1))
|
|
#
|
|
#
|
|
# dev.off()
|
|
|
|
################################################################
|
|
|
|
#####################################################################
|
|
#============
|
|
# Plot labels
|
|
#============
|
|
tit1 = "Stability outcome"
|
|
tit2 = "Affinity outcome"
|
|
tit3 = "Conservation outcome"
|
|
pt_size = 30
|
|
|
|
|
|
theme_georgia <- function(...) {
|
|
theme_gray(base_family = "sans", ...) +
|
|
theme(plot.title = element_text(face = "bold"))
|
|
}
|
|
|
|
|
|
title_theme <- calc_element("plot.title", theme_georgia())
|
|
|
|
pt1 = ggdraw() +
|
|
draw_label(
|
|
tit1,
|
|
fontfamily = title_theme$family,
|
|
fontface = title_theme$face,
|
|
#size = title_theme$size
|
|
size = pt_size
|
|
)
|
|
|
|
pt2 = ggdraw() +
|
|
draw_label(
|
|
tit2,
|
|
fontfamily = title_theme$family,
|
|
fontface = title_theme$face,
|
|
size = pt_size
|
|
)
|
|
|
|
pt3 = ggdraw() +
|
|
draw_label(
|
|
tit3,
|
|
fontfamily = title_theme$family,
|
|
fontface = title_theme$face,
|
|
size = pt_size
|
|
)
|
|
|
|
# extract common legend
|
|
common_legend_outcome = get_legend(mLigP +
|
|
guides(color = guide_legend(nrow = 1)) +
|
|
theme(legend.position = "top"))
|
|
|
|
|
|
|
|
my_label_size = 25
|
|
#======================
|
|
# Output plot function
|
|
#======================
|
|
OutPlotBP = function(x){
|
|
cowplot::plot_grid(
|
|
cowplot::plot_grid(pt1,
|
|
common_legend_outcome,
|
|
cowplot::plot_grid( duetP, foldxP
|
|
, deepddgP, dynamut2P
|
|
, nrow = 2
|
|
, ncol = 2
|
|
, labels = c("A", "B", "C","D")
|
|
, label_size = my_label_size
|
|
)
|
|
, ncol = 1
|
|
, rel_heights = c(7, 3, 90)),
|
|
|
|
cowplot::plot_grid(pt2,
|
|
cowplot::plot_grid(mLigP, mmLigP, ppi2P
|
|
, nrow = 1
|
|
, ncol = 3
|
|
, labels = c("E","F", "G")
|
|
, label_size = my_label_size
|
|
)
|
|
, ncol = 1
|
|
, rel_heights = c(1, 9)),
|
|
|
|
cowplot::plot_grid(pt3,
|
|
cowplot::plot_grid(consurfP, proveanP, snap2P
|
|
, nrow = 1
|
|
, ncol = 3
|
|
, labels = c("H", "I", "J")
|
|
, labels_x = 0.2
|
|
, label_size = my_label_size
|
|
, rel_widths = c(0.2, 0.2, 0.2)
|
|
)
|
|
, ncol = 1
|
|
, rel_heights = c(0.07, 0.93)
|
|
),
|
|
|
|
nrow = 3,
|
|
rel_heights = c(0.58, 0.25, 0.27),
|
|
align = "hv"
|
|
)
|
|
}
|
|
|
|
#=====================
|
|
# OutPlot: svg and png
|
|
#======================
|
|
#ratio 11.69 by 8.27
|
|
w = 8.27*2
|
|
h = 11.69*2
|
|
|
|
#svg
|
|
bp_all_CLP = paste0(outdir_images
|
|
,tolower(gene)
|
|
,"_bp_all_CL.svg")
|
|
cat(paste0("plot filename:", bp_all_CLP))
|
|
|
|
svg(bp_all_CLP, width = w, height = h)
|
|
OutPlotBP()
|
|
dev.off()
|
|
|
|
#png
|
|
bp_all_CLP_png = paste0(outdir_images
|
|
,tolower(gene)
|
|
,"_bp_all_CL.png")
|
|
cat(paste0("plot filename:", bp_all_CLP_png))
|
|
|
|
png(bp_all_CLP_png, width = w, height = h, units = "in", res = 300 )
|
|
OutPlotBP()
|
|
dev.off()
|
|
|
|
#####################################################################
|
|
# test
|
|
#
|
|
# setDT(df3)[, pos_count2 := .N, by = .(eval(parse(text = "position")))]
|
|
# foo = df3[, c("mutationinformation", "position")]
|
|
# df4 = foo[, c("mutationinformation", "position")]
|
|
#
|
|
#
|
|
# var_pos = "position"
|
|
# df4 =
|
|
# df4 %>%
|
|
# dplyr::add_count(eval(parse(text = var_pos)))
|
|
#
|
|
# class(df4)
|
|
# df4 = as.data.frame(df4)
|
|
# class(df4)
|
|
# nc_change = which(colnames(df4) == "n")
|
|
# colnames(df4)[nc_change] <- "pos_count"
|
|
# class(df4)
|
|
#
|
|
# setDT(df4)[, pos_count2 := .N, by = .(eval(parse(text = "position")))]
|
|
# class(df4)
|
|
#
|
|
# all(df4$pos_count==df4$pos_count2)
|
|
#
|
|
# # %>%
|
|
# #group_by(pos_count = position)
|
|
#
|
|
# # df4 =
|
|
# # df4 %>%
|
|
# # dplyr::group_by(position) %>%
|
|
# # count(position)
|
|
#foo2 = df4[, c("mutationinformation", "position", "pos_count")]
|
|
|
|
#####################################################################
|
|
# ------------------------------
|
|
# bp site site count: ALL
|
|
# <10 Ang ligand
|
|
# ------------------------------
|
|
posC_all = site_snp_count_bp(plotdf = df3
|
|
, df_colname = "position"
|
|
, xaxis_title = "Number of nsSNPs"
|
|
, yaxis_title = "Number of Sites"
|
|
, subtitle_size = 20)
|
|
|
|
# ------------------------------
|
|
# bp site site count: mCSM-lig
|
|
# < 10 Ang ligand
|
|
# ------------------------------
|
|
common_bp_title = paste0("Sites <", DistCutOff, angstroms_symbol)
|
|
|
|
posC_lig = site_snp_count_bp(plotdf = df3_lig
|
|
, df_colname = "position"
|
|
, xaxis_title = "Number of nsSNPs"
|
|
, yaxis_title = "Number of Sites"#+ annotate("text", x = 1.5, y = 2.2, label = "Text No. 1")
|
|
#, subtitle_text = paste0(common_bp_title, " ligand")
|
|
, subtitle_size = 8
|
|
, subtitle_colour = subtitle_colour)
|
|
posC_lig
|
|
# ------------------------------
|
|
# bp site site count: ppi2
|
|
# < 10 Ang interface
|
|
# ------------------------------
|
|
|
|
posC_ppi2 = site_snp_count_bp(plotdf = df3_ppi2
|
|
, df_colname = "position"
|
|
, xaxis_title = "Number of nsSNPs"
|
|
, yaxis_title = "Number of Sites"
|
|
, subtitle_text = paste0(common_bp_title, " interface")
|
|
, subtitle_size = 20
|
|
, subtitle_colour = subtitle_colour)
|
|
posC_ppi2
|
|
# ------------------------------
|
|
#FIXME: bp site site count: na
|
|
# < 10 Ang TBC
|
|
# ------------------------------
|
|
# posC_na = site_snp_count_bp(plotdf = df3_na
|
|
# , df_colname = "position"
|
|
# , xaxis_title = ""
|
|
# , yaxis_title = "")
|
|
|
|
|
|
#===========================
|
|
# output: SITE SNP count:
|
|
# all + affinity
|
|
#==========================
|
|
# my_label_size = 25
|
|
# pos_count_combined_CLP = paste0(outdir_images
|
|
# ,tolower(gene)
|
|
# ,"_pos_count_PS_AFF.svg")
|
|
#
|
|
#
|
|
# svg(pos_count_combined_CLP, width = 20, height = 5.5)
|
|
# print(paste0("plot filename:", pos_count_combined_CLP))
|
|
#
|
|
# cowplot::plot_grid(posC_all, posC_lig, posC_ppi2
|
|
# #, posC_na
|
|
# , nrow = 1
|
|
# , ncol = 3
|
|
# , labels = "AUTO"
|
|
# , label_size = my_label_size)
|
|
#
|
|
# dev.off()
|
|
|
|
|
|
#===============================================================
|