moved coloured_bp_data.R to redundant in light of updated function and reflected this in notes withing get_plotting_dfs.R

scripts/plotting/redundant/coloured_bp_data.R

@@ -0,0 +1,80 @@
+#!/usr/bin/env Rscript  
+#################################################################
+# TASK: Script to add bp colours ~ barplot heatmap
+#################################################################
+
+my_df = merged_df3
+
+cols_to_select = c("mutationinformation", "drtype"
+                   , "wild_type"
+                   , "position"
+                   , "mutant_type"
+                   , "chain", "ligand_id", "ligand_distance"
+                   , "duet_stability_change", "duet_outcome", "duet_scaled"
+                   , "ligand_affinity_change", "ligand_outcome", "affinity_scaled"
+                   , "ddg_foldx", "foldx_scaled", "foldx_outcome"
+                   , "deepddg", "deepddg_outcome" # comment out as not available for pnca
+                   , "asa", "rsa", "rd_values", "kd_values"
+                   , "af", "or_mychisq", "pval_fisher" 
+                   , "or_fisher", "or_logistic", "pval_logistic"
+                   , "wt_prop_water", "mut_prop_water", "wt_prop_polarity", "mut_prop_polarity"
+                   , "wt_calcprop", "mut_calcprop")
+
+#=======================
+# Data for sub colours
+# barplot: PS
+#=======================
+
+cat("\nNo. of cols to select:", length(cols_to_select))
+
+subcols_df_ps = my_df[, cols_to_select]
+
+cat("\nNo of unique positions for ps:"
+    , length(unique(subcols_df_ps$position)))
+
+# add count_pos col that counts the no. of nsSNPS at a position
+setDT(subcols_df_ps)[, pos_count := .N, by = .(position)]
+
+# should be a factor
+if (is.factor(subcols_df_ps$duet_outcome)){
+  cat("\nDuet_outcome is factor")
+  table(subcols_df_ps$duet_outcome)
+}else{
+  cat("\nConverting duet_outcome to factor")
+  subcols_df_ps$duet_outcome = as.factor(subcols_df_ps$duet_outcome)
+  table(subcols_df_ps$duet_outcome)
+}
+
+# should be -1 and 1
+min(subcols_df_ps$duet_scaled)
+max(subcols_df_ps$duet_scaled)
+
+tapply(subcols_df_ps$duet_scaled, subcols_df_ps$duet_outcome, min)
+tapply(subcols_df_ps$duet_scaled, subcols_df_ps$duet_outcome, max)
+
+# check unique values in normalised data
+cat("\nNo. of unique values in duet scaled, no rounding:"
+    , length(unique(subcols_df_ps$duet_scaled)))
+
+# No rounding    
+my_grp = subcols_df_ps$duet_scaled; length(my_grp)
+
+# Add rounding is to be used
+n = 3 
+subcols_df_ps$duet_scaledR = round(subcols_df_ps$duet_scaled, n)
+
+cat("\nNo. of unique values in duet scaled", n, "places rounding:"
+    , length(unique(subcols_df_ps$duet_scaledR)))
+
+my_grp_r = subcols_df_ps$duet_scaledR  # rounding
+
+# Add grp cols
+subcols_df_ps$group <- paste0(subcols_df_ps$duet_outcome, "_", my_grp, sep = "")
+subcols_df_ps$groupR <- paste0(subcols_df_ps$duet_outcome, "_", my_grp_r, sep = "")
+
+# Call the function to create the palette based on the group defined above
+subcols_ps <- ColourPalleteMulti(subcols_df_ps, "duet_outcome", "my_grp")
+subcolsR_ps <- ColourPalleteMulti(subcols_df_ps, "duet_outcome", "my_grp_r")
+
+cat("Colour palette generated for my_grp: ", length(subcols_ps), " colours")
+cat("Colour palette generated for my_grp_r: ", length(subcolsR_ps), " colours")