added output tables with active site

scripts/plotting/get_plotting_dfs.R

@@ -224,8 +224,8 @@ cat(s3)
 # 1
 # s3 = c("\nSuccessfully sourced lineage_data.R")
 # cat(s3)
- merged_df3 = as.data.frame(merged_df3)
+merged_df3 = as.data.frame(merged_df3)
-# 
+ 
 corr_df_m3_f = corr_data_extract(merged_df3
                                  , gene = gene
                                  , drug = drug

scripts/plotting/plotting_thesis/gid/gg_pairs_all_gid.R

@@ -1,7 +1,15 @@
-#source("~/git/LSHTM_analysis/config/embb.R")
+#=======
-#source("~/git/LSHTM_analysis/scripts/plotting/plotting_colnames.R")
+# Input
-#source("~/git/LSHTM_analysis/scripts/plotting/get_plotting_dfs.R")
+#=======
+source("~/git/LSHTM_analysis/config/gid.R")
+source("~/git/LSHTM_analysis/scripts/plotting/get_plotting_dfs.R")
+#=======
+# output
+#=======
+outdir_images = paste0("~/git/Writing/thesis/images/results/", tolower(gene), "/")
+cat("\nOutput dir for plots:", outdir_images)
 
+#### corr plot function: ggpairs ####
 my_gg_pairs=function(plot_df, plot_title
                      , tt_args_size = 2.5
                      , gp_args_size = 2.5){
@@ -38,8 +46,10 @@ my_gg_pairs=function(plot_df, plot_title
   #                           face="bold"))
 }
 
+
+#### Data for ggpairs ####
 DistCutOff = 10
-###########################################################################
+
 geneL_normal  = c("pnca")
 geneL_na      = c("gid", "rpob")
 geneL_ppi2    = c("alr", "embb", "katg", "rpob")
@@ -53,7 +63,7 @@ corr_plotdf = corr_data_extract(merged_df3
 
 aff_dist_cols = colnames(corr_plotdf)[grep("Dist", colnames(corr_plotdf))]
 static_cols =  c("Log10(MAF)"
-                 , "Log10(OR)"
+                 #, "Log10(OR)"
 )
 ############################################################
 #=============================================
@@ -148,7 +158,7 @@ plot_corr_df_aff = my_gg_pairs(corr_df_aff
                                #, gp_args_size = 4
                                )
 
-#### Combine plots #####
+#### Combine plots (OLD, AB + C) #####
 # #png("/home/tanu/tmp/gg_pairs_all.png", height = 6, width=11.75, unit="in",res=300)
 # png(paste0(outdir_images
 #            ,tolower(gene)

scripts/plotting/plotting_thesis/gid/gid_ORandSNP_results.R

@@ -136,8 +136,6 @@ bar = bar[, c("mutationinformation"
 
 table(bar$sensitivity)
 
-table(bar$or_mychisq>1&bar$signif_fdr) # sen and res ~ OR
-
 str(bar)
 sen = bar[bar$or_mychisq<1,]
 sen = na.omit(sen)
@@ -162,7 +160,9 @@ if (nrow(bar_or) == nrow(sen1) + nrow(res1) ){
 
 # percent for OR muts
 pc_orR = nrow(res1)/(nrow(sen1) + nrow(res1)); pc_orR
-cat("\nPercentage of muts with OR>1 i.e resistant:"
+
+cat("Number of R muts with OR>1:", nrow(res1)
+  , "\nPercentage of muts with OR>1 i.e resistant:"
     , pc_orR *100 )
 
 # muts with highest OR
@@ -171,22 +171,23 @@ head(bar_or$mutationinformation, 10)
 # sort df
 bar_or = bar_or[order(bar_or$or_mychisq
                       , bar_or$ligand_distance
-                      , bar_or$interface_dist
+                      , bar_or$nca_dist
+                      #, bar_or$interface_dist
                       , decreasing = T), ]
 
+nrow(bar_or)
 bar_or$drug_site = ifelse(bar_or$position%in%aa_pos_drug, "drug", "no")
 table(bar_or$drug_site)
 
-bar_or$dsl_site = ifelse(bar_or$position%in%aa_pos_dsl, "dsl", "no")
+bar_or$rna_site = ifelse(bar_or$position%in%aa_pos_rna, "rna", "no")
 table(bar_or$dsl_site)
 
-bar_or$ca_site = ifelse(bar_or$position%in%aa_pos_ca, "ca", "no")
+bar_or$sam_site = ifelse(bar_or$position%in%aa_pos_sam, "sam", "no")
 table(bar_or$ca_site)
 
-bar_or$cdl_site = ifelse(bar_or$position%in%aa_pos_cdl, "cdl", "no")
+bar_or$amp_site = ifelse(bar_or$position%in%aa_pos_amp, "amp", "no")
 table(bar_or$cdl_site)
 
-
 top10_or = bar_or[1:10,]
 
 # are these active sites

scripts/plotting/plotting_thesis/gid/gid_appendix_tables.R

@@ -285,6 +285,7 @@ if (identical(colnames(mut_h_avs_dd), colnames(mut_h_avs_ss)) ){
 #-------------------
 # Filtered columns
 # most DD/SS: ligand
+# FIXME DUBIOUS as min and max can be both negative
 #-------------------
 df3_effects_lig = df3_effects[df3_effects[[LigDist_colname]]<DistCutOff,]
 nrow(df3_effects_lig)
@@ -296,7 +297,7 @@ if (identical(colnames(mut_h_lig_dd), colnames(mut_h_lig_ss)) ){
   
   # add cols
   mut_h_lig_dd$mutational_effect = "Most Destabilising for Ligand affinity"
-  mut_h_lig_ss$mutational_effect = "Most Stabilising for Ligand affinity"
+  mut_h_lig_ss$mutational_effect = "CAUTION: Most DE/Stabilising for Ligand affinity"
   
   cat("\nPass: avg ligand affinity")
 }else{
@@ -415,6 +416,20 @@ if ( length(colnames(combined_table)) == length(colsNames_combined_table) ) {
   stop("\nAbort: No. of cols mismatch. Cannot assign pretty colnames for output")
 }
 
+
+nrow(combined_table)
+combined_table$drug_site = ifelse(combined_table$position%in%aa_pos_drug, "drug", "no")
+table(combined_table$drug_site)
+
+combined_table$rna_site = ifelse(combined_table$position%in%aa_pos_rna, "rna", "no")
+table(combined_table$rna_site)
+
+combined_table$sam_site = ifelse(combined_table$position%in%aa_pos_sam, "sam", "no")
+table(combined_table$sam_site)
+
+combined_table$amp_site = ifelse(combined_table$position%in%aa_pos_amp, "amp", "no")
+table(combined_table$amp_site)
+
 #--------------------
 # write output file: KS test within grpup
 #----------------------