Split out simple vars in interaction scores

workflow/apps/r/calculate_interaction_zscores5.R

@@ -230,17 +230,20 @@ calculate_interaction_scores <- function(df, max_conc, variables, group_vars = c
   bg_sd_r <- df %>% filter(conc_num_factor == 0) %>% pull(sd_r) %>% first()
   bg_sd_AUC <- df %>% filter(conc_num_factor == 0) %>% pull(sd_AUC) %>% first()
   
-  # Calculate interaction scores
+  # First mutate block to calculate NG, DB, SM
   print("Calculating interaction scores part 1")
   interaction_scores <- df %>%
     group_by(across(all_of(group_vars)), conc_num, conc_num_factor) %>%
     mutate(
       NG = sum(NG, na.rm = TRUE),
       DB = sum(DB, na.rm = TRUE),
-      SM = sum(SM, na.rm = TRUE)
-    ) %>%
+      SM = sum(SM, na.rm = TRUE),
+      N = ~sum(!is.na(L)),  # Count of non-NA values
+    )
+  
+  # Second mutate block to calculate variables and Delta using NG
+  interaction_scores <- interaction_scores %>%
     mutate(across(all_of(variables), list(
-      N = ~sum(!is.na(.)),  # Count of non-NA values
       mean = ~mean(., na.rm = TRUE),
       median = ~median(., na.rm = TRUE),
       max = ~max(., na.rm = TRUE),
@@ -269,7 +272,7 @@ calculate_interaction_scores <- function(df, max_conc, variables, group_vars = c
       Delta_L = if_else(SM == 1, mean_L - WT_L, Delta_L) # disregard shift for set to max values in Z score calculation
     ) %>%
     ungroup()
-  
+
   # Calculate linear models and interaction scores per gene
   print("Calculating interaction scores part 2")
   interaction_scores_all <- interaction_scores %>%
@@ -296,6 +299,7 @@ calculate_interaction_scores <- function(df, max_conc, variables, group_vars = c
   return(list(zscores_calculations = interaction_scores_all, zscores_interactions = interaction_scores))
 }
 
+
 interaction_plot_configs <- function(df, variable) {
   ylim_vals <- switch(variable,
     "L" = c(-65, 65),
@@ -715,8 +719,6 @@ main <- function() {
       reference_strain <- process_strains(df_reference) # TODO double-check
       deletion_strains <- process_strains(df_deletion) # TODO double-check
       
-      # TODO we may need to add "num" to grouping vars
-      
       # Calculate interactions
       variables <- c("L", "K", "r", "AUC")
       # We are recalculating some of the data here