268 lines
9.4 KiB
R
268 lines
9.4 KiB
R
#!/usr/bin/env Rscript
|
|
|
|
#########################################################
|
|
# TASK: Get formatted data for plots
|
|
#########################################################
|
|
# working dir and loading libraries
|
|
getwd()
|
|
source("~/git/LSHTM_analysis/scripts/Header_TT.R")
|
|
source("~/git/LSHTM_analysis/scripts/plotting/plotting_colnames.R")
|
|
# cmd args passed
|
|
# in from other scripts
|
|
# to call this
|
|
# set drug and gene name
|
|
|
|
#==========================================
|
|
# variables for affinity:
|
|
# comes from functions/plotting_globals.R
|
|
#==========================================
|
|
|
|
cat("\nGlobal variables for Ligand:"
|
|
, "\nligand distance colname:", LigDist_colname
|
|
, "\nligand distance cut off:", LigDist_cutoff)
|
|
|
|
cat("\nGlobal variables for mCSM-PPI2 affinity:"
|
|
, "\nPPI2 distance colname:", ppi2Dist_colname
|
|
, "\nPPI2 cut off:", DistCutOff)
|
|
|
|
cat("\nGlobal variables for mCSM-NA affinity:"
|
|
, "\nligand distance colname:", naDist_colname
|
|
, "\nligand distance cut off:", DistCutOff)
|
|
|
|
|
|
#===========
|
|
# input
|
|
#===========
|
|
#--------------------------------------------
|
|
# call: import_dirs()
|
|
# comes from functions/plotting_globals.R
|
|
#--------------------------------------------
|
|
import_dirs(drug, gene)
|
|
|
|
#---------------------------
|
|
# call: plotting_data()
|
|
#---------------------------
|
|
if (!exists("infile_params") && exists("gene")){
|
|
#if (!is.character(infile_params) && exists("gene")){ # when running as cmd
|
|
in_filename_params = paste0(tolower(gene), "_all_params.csv")
|
|
infile_params = paste0(outdir, "/", in_filename_params)
|
|
cat("\nInput file for mcsm comb data not specified, assuming filename: ", infile_params, "\n")
|
|
}
|
|
|
|
# Input 1:
|
|
cat("\nReading mcsm combined data file: ", infile_params)
|
|
mcsm_df = read.csv(infile_params, header = T)
|
|
if (tolower(gene)%in%c("rpob")){
|
|
mcsm_df = mcsm_df[mcsm_df$position!=1148,]
|
|
}
|
|
|
|
pd_df = plotting_data(mcsm_df
|
|
, gene = gene # ADDED
|
|
, lig_dist_colname = LigDist_colname
|
|
, lig_dist_cutoff = LigDist_cutoff)
|
|
|
|
my_df = pd_df[[1]]
|
|
my_df_u = pd_df[[2]] # this forms one of the input for combining_dfs_plotting()
|
|
|
|
max_ang <- round(max(my_df_u[[LigDist_colname]]))
|
|
min_ang <- round(min(my_df_u[[LigDist_colname]]))
|
|
|
|
cat("\nLigand distance colname:", LigDist_colname
|
|
, "\nThe max distance", gene, "structure df" , ":", max_ang, "\u212b"
|
|
, "\nThe min distance", gene, "structure df" , ":", min_ang, "\u212b")
|
|
|
|
#--------------------------------
|
|
# call: combining_dfs_plotting()
|
|
#--------------------------------
|
|
if (!exists("infile_metadata") && exists("gene")){
|
|
#if (!is.character(infile_metadata) && exists("gene")){ # when running as cmd
|
|
in_filename_metadata = paste0(tolower(gene), "_metadata.csv") # part combined for gid
|
|
infile_metadata = paste0(outdir, "/", in_filename_metadata)
|
|
cat("\nInput file for gene metadata not specified, assuming filename: ", infile_metadata, "\n")
|
|
}
|
|
|
|
# Input 2: read <gene>_meta data.csv
|
|
cat("\nReading meta data file: ", infile_metadata)
|
|
|
|
gene_metadata <- read.csv(infile_metadata
|
|
, stringsAsFactors = F
|
|
, header = T)
|
|
|
|
cat("\nDim of meta data file: ", dim(gene_metadata))
|
|
|
|
all_plot_dfs = combining_dfs_plotting(my_df_u
|
|
, gene_metadata
|
|
#, gene = gene # ADDED
|
|
, lig_dist_colname = LigDist_colname
|
|
, lig_dist_cutoff = LigDist_cutoff)
|
|
|
|
merged_df2 = all_plot_dfs[[1]]
|
|
merged_df3 = all_plot_dfs[[2]]
|
|
#merged_df2_comp = all_plot_dfs[[3]]
|
|
#merged_df3_comp = all_plot_dfs[[4]]
|
|
|
|
#======================================================================
|
|
|
|
####################################################################
|
|
# Data for subcols barplot (~heatmap)
|
|
####################################################################
|
|
|
|
#source("coloured_bp_data.R")
|
|
# Repurposed function so that params can be passed instead to generate
|
|
# data required for plotting.
|
|
# Moved "coloured_bp_data.R" to redundant/
|
|
|
|
####################################################################
|
|
# Data for logoplots
|
|
####################################################################
|
|
#
|
|
# source(paste0(plot_script_path, "logo_data_msa.R"))
|
|
# s1 = c("\nSuccessfully sourced logo_data_msa.R")
|
|
# cat(s1)
|
|
#
|
|
# ####################################################################
|
|
# # Data for DM OM Plots: WF and LF dfs
|
|
# # My function: dm_om_wf_lf_data()
|
|
# # location: scripts/functions/dm_om_data.R
|
|
# #source("other_plots_data.R")
|
|
# ####################################################################
|
|
#
|
|
# #source(paste0(plot_script_path, "dm_om_data.R")) # calling the function directly instead
|
|
geneL_normal = c("pnca")
|
|
geneL_na = c("gid", "rpob")
|
|
geneL_ppi2 = c("alr", "embb", "katg", "rpob")
|
|
|
|
# geneL_normal = c("pnca")
|
|
# geneL_both = c("rpob")
|
|
# geneL_ppi2 = c("alr", "embb", "katg")
|
|
# geneL_na = c("gid")
|
|
|
|
all_dm_om_df = dm_om_wf_lf_data(df = merged_df3, gene = gene)
|
|
|
|
wf_duet = all_dm_om_df[['wf_duet']]
|
|
lf_duet = all_dm_om_df[['lf_duet']]
|
|
|
|
wf_mcsm_lig = all_dm_om_df[['wf_mcsm_lig']]
|
|
lf_mcsm_lig = all_dm_om_df[['lf_mcsm_lig']]
|
|
|
|
wf_foldx = all_dm_om_df[['wf_foldx']]
|
|
lf_foldx = all_dm_om_df[['lf_foldx']]
|
|
|
|
wf_deepddg = all_dm_om_df[['wf_deepddg']]
|
|
lf_deepddg = all_dm_om_df[['lf_deepddg']]
|
|
|
|
wf_dynamut2 = all_dm_om_df[['wf_dynamut2']]
|
|
lf_dynamut2 = all_dm_om_df[['lf_dynamut2']]
|
|
|
|
wf_consurf = all_dm_om_df[['wf_consurf']]
|
|
lf_consurf = all_dm_om_df[['lf_consurf']]
|
|
|
|
wf_snap2 = all_dm_om_df[['wf_snap2']]
|
|
lf_snap2 = all_dm_om_df[['lf_snap2']]
|
|
|
|
wf_provean = all_dm_om_df[['wf_provean']]
|
|
lf_provean = all_dm_om_df[['lf_provean']]
|
|
|
|
# NEW
|
|
wf_dist_gen = all_dm_om_df[['wf_dist_gen']]
|
|
lf_dist_gen = all_dm_om_df[['lf_dist_gen']]
|
|
|
|
# ppi2 genes
|
|
if (tolower(gene)%in%geneL_ppi2){
|
|
wf_mcsm_ppi2 = all_dm_om_df[['wf_mcsm_ppi2']]
|
|
lf_mcsm_ppi2 = all_dm_om_df[['lf_mcsm_ppi2']]
|
|
}
|
|
|
|
# na genes
|
|
if (tolower(gene)%in%geneL_na){
|
|
wf_mcsm_na = all_dm_om_df[['wf_mcsm_na']]
|
|
lf_mcsm_na = all_dm_om_df[['lf_mcsm_na']]
|
|
}
|
|
|
|
# both ppi2+na genes:: NOT NEEDED Here as its is handled by the two ifs above
|
|
# if (tolower(gene)%in%geneL_both){
|
|
# wf_mcsm_ppi2 = all_dm_om_df[['wf_mcsm_ppi2']]
|
|
# lf_mcsm_ppi2 = all_dm_om_df[['lf_mcsm_ppi2']]
|
|
#
|
|
# wf_mcsm_na = all_dm_om_df[['wf_mcsm_na']]
|
|
# lf_mcsm_na = all_dm_om_df[['lf_mcsm_na']]
|
|
# }
|
|
|
|
|
|
s2 = c("\nSuccessfully sourced other_plots_data.R")
|
|
cat(s2)
|
|
#
|
|
# ####################################################################
|
|
# # Data for Lineage barplots: WF and LF dfs
|
|
# # My function: lineage_plot_data()
|
|
# # location: scripts/functions/lineage_plot_data.R
|
|
# ####################################################################
|
|
#
|
|
source(paste0(plot_script_path, "lineage_data.R"))
|
|
# # converted to a function. Moved lineage_data.R to redundant/
|
|
lineage_dfL = lineage_plot_data(merged_df2
|
|
, lineage_column_name = "lineage"
|
|
, remove_empty_lineage = F
|
|
, lineage_label_col_name = "lineage_labels"
|
|
, id_colname = "id"
|
|
, snp_colname = "mutationinformation"
|
|
)
|
|
|
|
lin_wf = lineage_dfL[['lin_wf']]
|
|
lin_lf = lineage_dfL[['lin_lf']]
|
|
|
|
s3 = c("\nSuccessfully sourced lineage_data.R")
|
|
cat(s3)
|
|
|
|
####################################################################
|
|
# Data for corr plots:
|
|
# My function: corr_data_extract()
|
|
# location: scripts/functions/corr_plot_data.R
|
|
####################################################################
|
|
# make sure the above script works because merged_df2_combined is needed
|
|
merged_df3 = as.data.frame(merged_df3)
|
|
|
|
corr_df_m3_f = corr_data_extract(merged_df3
|
|
, gene = gene
|
|
, drug = drug
|
|
, extract_scaled_cols = F)
|
|
head(corr_df_m3_f)
|
|
|
|
# corr_df_m2_f = corr_data_extract(merged_df2
|
|
# , gene = gene
|
|
# , drug = drug
|
|
# , extract_scaled_cols = F)
|
|
# head(corr_df_m2_f)
|
|
|
|
s4 = c("\nSuccessfully sourced Corr_data.R")
|
|
cat(s4)
|
|
|
|
########################################################################
|
|
# End of script
|
|
########################################################################
|
|
# if ( all( length(s1), length(s2), length(s3), length(s4) ) > 0 ){
|
|
# cat(
|
|
# "\n##################################################"
|
|
# , "\nSuccessful: get_plotting_dfs.R worked!"
|
|
# , "\n###################################################\n")
|
|
# } else {
|
|
# cat(
|
|
# "\n#################################################"
|
|
# , "\nFAIL: get_plotting_dfs.R didn't complete fully!Please check"
|
|
# , "\n###################################################\n" )
|
|
# }
|
|
#
|
|
########################################################################
|
|
# clear excess variables: from the global enviornment
|
|
|
|
vars0 = ls(envir = .GlobalEnv)[grepl("curr_*", ls(envir = .GlobalEnv))]
|
|
vars1 = ls(envir = .GlobalEnv)[grepl("^cols_to*", ls(envir = .GlobalEnv))]
|
|
vars2 = ls(envir = .GlobalEnv)[grepl("pivot_cols_*", ls(envir = .GlobalEnv))]
|
|
vars3 = ls(envir = .GlobalEnv)[grepl("expected_*", ls(envir = .GlobalEnv))]
|
|
|
|
rm( infile_metadata
|
|
, infile_params
|
|
, vars0
|
|
, vars1
|
|
, vars2
|
|
, vars3)
|