repurposing corr_data.R into a function to allow required params to be passed in

scripts/Header_TT.R

@@ -8,11 +8,35 @@
 
 require("getopt", quietly = TRUE) # cmd parse arguments
 
+if (!require ("DT")){
+  install.packages("DT")
+  library(DT)
+}
+
+if (!require ("plyr")){
+  install.packages("plyr")
+  library(plyr)
+}
+
 if (!require("tidyverse")) {
   install.packages("tidyverse", dependencies = TRUE)
   library(tidyverse)
 }
 
+#---------------------------
+# covered by tidyverse
+
+# if (!require("ggplot2")) {
+#   install.packages("ggplot2", dependencies = TRUE)
+#   library(ggplot2)
+# }
+
+# if (!require ("dplyr")){
+#    install.packages("dplyr")
+#    library(dplyr)
+#  }
+#-----------------------------
+
 if (!require("shiny")) {
   install.packages("shiny", dependencies = TRUE)
   library(shiny)
@@ -33,26 +57,6 @@ if (!require("ggridges")) {
   library(ggridges)
 }
 
-# if (!require("ggplot2")) {
-#   install.packages("ggplot2", dependencies = TRUE)
-#   library(ggplot2)
-# }
-
-# if (!require ("dplyr")){
-#   install.packages("dplyr")
-#   library(dplyr)
-# }
-
-if (!require ("DT")){
-  install.packages("DT")
-  library(DT)
-}
-
-if (!require ("plyr")){
-   install.packages("plyr")
-   library(plyr)
- }
-
 # Install
 #if(!require(devtools)) install.packages("devtools")
 #devtools::install_github("kassambara/ggcorrplot")
@@ -188,3 +192,17 @@ map(paste0(func_path, source_files), source)  # source all your R scripts!
 
 # set plot script dir
 plot_script_path = "~/git/LSHTM_analysis/scripts/plotting/"
+
+##################################################
+
+# Function name clashes with plyr and dplyr
+# # loading dplyr after plyr causes issues
+# if("dplyr" %in% (.packages())){
+#   detach("package:dplyr", unload=TRUE) 
+#   detach("package:plyr", unload=TRUE) 
+# } 
+# library(plyr)
+# library(dplyr)
+
+# another solution is to requireNamespace() instead of library()
+# so its function names don't collide with dplyr's.

scripts/plotting/corr_data.R

@@ -2,6 +2,7 @@
 #########################################################
 # TASK: Script to format data for corr plots
 #########################################################
+#library(dplyr)
 
 #=================================================
 #         Data for Corrplots
@@ -12,6 +13,10 @@ cat("\n=========================================="
 
 # use data
 #merged_df2
+geneL_normal = c("pnca")
+geneL_na_dy = c("gid")
+geneL_na = c("rpob")
+geneL_ppi2 = c("alr", "embb", "katg", "rpob")
 
 #----------------------------
 # columns for corr plots:PS
@@ -19,11 +24,55 @@ cat("\n=========================================="
 # NOTE: you can add mcsm_ppi column as well, and it will only select what it can find!
 big_df_colnames = data.frame(names(merged_df2))
 
-corr_cols_select <- c("mutationinformation", drug, "mutation_info_labels"
+core_cols = c("mutationinformation", drug, "mutation_info_labels"
               , "duet_stability_change", "ligand_affinity_change", "ddg_foldx", "asa", "rsa"
               , "rd_values", "kd_values", "log10_or_mychisq", "neglog_pval_fisher","af"
-                   , "deepddg", "ddg_dynamut", "ddg_dynamut2", "mcsm_na_affinity"
+              , "deepddg" , "ddg_dynamut2"
-                   , "ddg_encom", "dds_encom", "ddg_mcsm", "ddg_sdm", "ddg_duet", "ligand_distance")
+              , "consurf_score"
+              #, "consurf_scaled"
+              , "snap2_score"
+              #, "snap2_scaled", "snap2_accuracy_pc"
+              , "ligand_distance")
+
+if (tolower(gene)%in%geneL_normal){
+  corr_cols_select = core_cols
+}
+if (tolower(gene)%in%geneL_na_dy){
+  additional_cols = c("mcsm_na_affinity"
+                    , "ddg_dynamut"
+                    , "ddg_encom", "dds_encom"
+                    , "ddg_mcsm", "ddg_sdm"
+                    , "ddg_duet"
+                    #, "mcsm_na_scaled"
+                    #, "ddg_dynamut_scaled"
+                    #, "ddg_encom_scaled", "dds_encom_scaled"
+                    #, "ddg_mcsm_scaled", "ddg_sdm_scaled"
+                    #, "ddg_duet_scaled"
+                    )
+  
+  corr_cols_select = c(core_cols, additional_cols)
+  
+}
+
+if (tolower(gene)%in%geneL_na){
+  additional_cols = c("mcsm_na_affinity"
+                      #, "mcsm_na_scaled"
+                      )
+  
+  corr_cols_select = c(core_cols, additional_cols)
+  
+}
+
+if (tolower(gene)%in%geneL_ppi2){
+  additional_cols = c("mcsm_ppi2_affinity")
+  corr_cols_select = c(core_cols, additional_cols)
+}
+
+# corr_cols_select <- c("mutationinformation", drug, "mutation_info_labels"
+#                    , "duet_stability_change", "ligand_affinity_change", "ddg_foldx", "asa", "rsa"
+#                    , "rd_values", "kd_values", "log10_or_mychisq", "neglog_pval_fisher","af"
+#                    , "deepddg", "ddg_dynamut", "ddg_dynamut2", "mcsm_na_affinity"
+#                    , "ddg_encom", "dds_encom", "ddg_mcsm", "ddg_sdm", "ddg_duet", "ligand_distance")
 
 #===========================
 # Corr data for plots: PS
@@ -36,9 +85,8 @@ corr_df_m2 = merged_df2[,colnames(merged_df2)%in%corr_cols_select]
 # formatting: some cols
 # Add pretty colnames
 #-----------------------
-corr_df_m2_f <- corr_df_m2 %>% 
+corr_df_m2_f <- corr_df_m2 %>% dplyr::rename(
-  rename(
+      'DUET'       = duet_stability_change
-      DUET       = duet_stability_change
     , 'mCSM-lig' = ligand_affinity_change
     , FoldX      = ddg_foldx
     , DeepDDG    = deepddg

scripts/plotting/get_plotting_dfs.R

@@ -124,32 +124,42 @@ cat(s1)
 ####################################################################
 #                        Data for DM OM Plots: Long format dfs
 ####################################################################
-
 #source("other_plots_data.R")
 
 #source(paste0(plot_script_path, "dm_om_data.R"))
-# 
+ 
 #s2 = c("\nSuccessfully sourced other_plots_data.R")
 #cat(s2)
 
 ####################################################################
 #                  Data for Lineage barplots: WF and LF dfs
 ####################################################################
-# 
+ 
-# source(paste0(plot_script_path, "lineage_data.R"))
+source(paste0(plot_script_path, "lineage_data.R"))
-# 
+
-# s3 = c("\nSuccessfully sourced lineage_data.R")
+s3 = c("\nSuccessfully sourced lineage_data.R")
-# cat(s3)
+cat(s3)
 
 ####################################################################
 #                  Data for corr plots:
 ####################################################################
 # make sure the above script works because merged_df2_combined is needed
 #source(paste0(plot_script_path, "corr_data.R"))
-# 
+
 #s4 = c("\nSuccessfully sourced corr_data.R")
 #cat(s4)
 
+# Moved "logo_data.R" to redundant/
+#-----------------------------------------
+# Replaced above with my function
+# corr_data_extract()
+
+corr_df_m3_f = corr_data_extract(merged_df3, extract_scaled_cols = F)
+head(corr_df_m3_f)
+
+corr_df_m2_f = corr_data_extract(merged_df2, extract_scaled_cols = F)
+head(corr_df_m2_f)
+
 ########################################################################
 #                           End of script
 ########################################################################
@@ -164,7 +174,7 @@ cat(s1)
 #  , "\nFAIL: get_plotting_dfs.R didn't complete fully!Please check"
 #  , "\n###################################################\n" )
 # }   
-#  
+ 
 ########################################################################
 # clear excess variables: from the global enviornment
 

scripts/plotting/other_plots_data.R

@@ -6,7 +6,7 @@
 #=======================================================================
 # working dir and loading libraries
 getwd()
-setwd("~/git/LSHTM_analysis/scripts/plotting")
+#setwd("~/git/LSHTM_analysis/scripts/plotting")
 getwd()
 
 #source("~/git/LSHTM_analysis/scripts/Header_TT.R")
@@ -14,11 +14,11 @@ library(ggplot2)
 library(data.table)
 library(dplyr)
 library(tidyverse)
-source("combining_dfs_plotting.R")
+#source("combining_dfs_plotting.R")
 
-rm(merged_df2, merged_df2_comp, merged_df2_lig, merged_df2_comp_lig
+#rm(merged_df2, merged_df2_comp, merged_df2_lig, merged_df2_comp_lig
-   , merged_df3_comp, merged_df3_comp_lig
+#   , merged_df3_comp, merged_df3_comp_lig
-   , my_df_u, my_df_u_lig)
+#   , my_df_u, my_df_u_lig)
 
 
 cols_to_select = c("mutation", "mutationinformation"