small edits in fig + rename

extras/concept_mapping_coverage_per_domain.R → extras/conceptMappingCoveragePerDomain.R

@@ -57,7 +57,7 @@ write.csv(table, file="concept_mapping_coverage.csv", row.names = FALSE)
 
 # Coverage plot like fig 6 in EHDEN DoA
 # note: coverage is percentage NOT failing to map
-coverage_results %>%
+fig <- coverage_results %>%
   # To keep things simple, we only look at the six main domains and units
   filter(domainField %in% c("VISIT", "PROCEDURE", "DRUG", "CONDITION", "MEASUREMENT", 
                        "OBSERVATION", "MEAS-UNIT", "OBS-UNIT")
@@ -69,20 +69,19 @@ coverage_results %>%
   geom_col() +
   geom_text(aes(label=scales::percent(coveragePct, accuracy = 0.01)), 
             position=position_stack(vjust=0.5), 
-            size=3, colour="white", fontface="bold") +
+            size=3, colour="gray10", fontface="bold") +
+  theme_minimal() +
   theme(
     axis.text.y=element_text(size=10),
     strip.placement="outside",
-    strip.text.y=element_text(angle=90, hjust=0.5, face="bold", size=6,
-                              margin=margin(r=0))
+    strip.text.y=element_text(angle=0, hjust=0.5, face="bold", size=6)
   ) +
   coord_flip() +
   facet_grid(domainField ~ ., scales="free_y", space="free_y", switch="y") +
-  scale_y_continuous(labels = scales::percent) +
   guides(fill=FALSE) +
   ylab("Percentage Coverage (%)") + 
   xlab("") + 
-  scale_fill_manual(values=c("darkblue", "skyblue"))
+  scale_fill_manual(values=c("cornflowerblue", "skyblue"))
 
-ggsave('concept_mapping_coverage.png', height = 5, width = 5 * 1.61803)
+ggsave('concept_mapping_coverage.png', height = 8, width = 8 * 1.61803)
 

extras/edit_thresholds/edit_thresholds.R → extras/edit_thresholds/editThresholds.R

@@ -80,12 +80,12 @@ get_threshold_location <- function(df, row){
       filter(
         checkName == row$checkName &
         cdmTableName == row$cdmTableName)
-
+    
     # for Field+Concept Level only
     if(tolower(row$Level) %in% c("field", "concept")){
       df3<-df3%>%filter(cdmFieldName == row$cdmFieldName)
     }
-
+    
     # 'additional' auxiliary columns
     if(row$checkName=="isForeignKey"){
       df3<-df3%>%filter(fkTableName==row$fkTableName)}
@@ -94,7 +94,7 @@ get_threshold_location <- function(df, row){
     if(tolower(row$Level) == "concept" & 
       row$checkName%in%c("plausibleValueLow", "plausibleValueHigh")){
       df3 <- df3%>%filter(conceptId==row$conceptId &
-                        unitConceptId==row$unitConceptId)}
+                          unitConceptId==row$unitConceptId)}
     return(df3$index)
 }
 
@@ -111,37 +111,37 @@ edit_threshold <- function(df, index, row){
   notes_old <- stringr::str_replace_all(notes_old, "[|][|]", "")
   notes_new <- paste(notes_old, row$Notes, sep="|")
   df[index, 'Notes'] = notes_new
-
+  
   return(df)
 
 }
 
 main_function <- function(file, file_new, thresholds){
-    # 1) get longer table
-    df <- read.csv(file, colClasses = "character", stringsAsFactors = FALSE)
-    df_long <- pivot_longer_func(df)
-    
-    # 2) Edit thresholds
-    for(i in c(1:nrow(thresholds))){
-      row = thresholds[i,]
-      index <- get_threshold_location(df_long, row)
-      df_long <- edit_threshold(df_long, index, row)
-    }
-    
-    # 3) get originally wide table
-    df_new <- pivot_wider_func(df=df_long)
-    df_new <- df_new[colnames(df)]
-    
-    # 4) save it!
-    write.csv(df_new, file_new, row.names = FALSE)
-    
-    # 5) (optional) final check
-    df.old <- read.csv(file, colClasses = "character", stringsAsFactors = FALSE)
-    df.new <- read.csv(file_new, colClasses = "character", stringsAsFactors = FALSE)
-    end <- base::all.equal(df.old, df.new)
-    end <- stringr::str_replace(end, "string mismatch", "difference")
-    end <- stringr::str_replace(end, "TRUE", "No differences")
-    print(end)
+  # 1) get longer table
+  df <- read.csv(file, colClasses = "character", stringsAsFactors = FALSE)
+  df_long <- pivot_longer_func(df)
+
+  # 2) Edit thresholds
+  for(i in c(1:nrow(thresholds))){
+    row = thresholds[i,]
+    index <- get_threshold_location(df_long, row)
+    df_long <- edit_threshold(df_long, index, row)
+  }
+
+  # 3) get originally wide table
+  df_new <- pivot_wider_func(df=df_long)
+  df_new <- df_new[colnames(df)]
+
+  # 4) save it!
+  write.csv(df_new, file_new, row.names = FALSE)
+
+  # 5) (optional) final check
+  df.old <- read.csv(file, colClasses = "character", stringsAsFactors = FALSE)
+  df.new <- read.csv(file_new, colClasses = "character", stringsAsFactors = FALSE)
+  end <- base::all.equal(df.old, df.new)
+  end <- stringr::str_replace(end, "string mismatch", "difference")
+  end <- stringr::str_replace(end, "TRUE", "No differences")
+  print(end)
 }
 
 # Main function