ml df stuff

scripts/plotting/get_ml_dfs.R

@@ -0,0 +1,103 @@
+#!/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 = ''
+                                      , lig_dist_cutoff = ''
+                                      , plotting = FALSE)
+
+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]]