update

R/make_other.R

@@ -0,0 +1,65 @@
+require(dplyr)
+
+#' Title
+#'
+#' @param dataframe
+#' @param value
+#'
+#' @return
+#' @export
+#'
+#' @examples
+make_other <- function(dataframe, value = "peak_area") {
+  top_4 <- dataframe |>
+    dplyr::group_by(best_candidate_1, best_candidate_2) |>
+    dplyr::mutate(new = sum(!!as.name(value))) |>
+    dplyr::distinct(new, .keep_all = TRUE) |>
+    dplyr::group_by(best_candidate_1) |>
+    dplyr::slice_max(new, n = 4, with_ties = FALSE) |>
+    dplyr::ungroup() |>
+    dplyr::select(
+      smiles_2D,
+      inchikey_2D,
+      best_candidate_1,
+      best_candidate_2,
+      best_candidate_3
+    ) |>
+    dplyr::distinct()
+
+  last <- dataframe |>
+    dplyr::ungroup() |>
+    dplyr::anti_join(
+      top_4,
+      by = c(
+        "best_candidate_1" = "best_candidate_1",
+        "best_candidate_2" = "best_candidate_2"
+      )
+    ) |>
+    dplyr::mutate(best_candidate_2 = "Other") |>
+    dplyr::select(
+      smiles_2D,
+      inchikey_2D,
+      best_candidate_1,
+      best_candidate_2,
+      best_candidate_3
+    ) |>
+    dplyr::distinct()
+
+  new <- dplyr::bind_rows(
+    top_4,
+    last
+  ) |>
+    dplyr::select(
+      -smiles_2D,
+      -inchikey_2D
+    )
+
+  df_new <- dataframe |>
+    dplyr::select(
+      -best_candidate_1,
+      -best_candidate_2
+    ) |>
+    dplyr::left_join(new)
+
+  return(df_new)
+}

R/no_other.R

@@ -0,0 +1,20 @@
+require(dplyr)
+
+#' Title
+#'
+#' @param dataframe
+#'
+#' @return
+#' @export
+#'
+#' @examples
+no_other <- function(dataframe) {
+  dataframe_no_other <- dataframe |>
+    dplyr::filter(!grepl(
+      pattern = "not",
+      x = best_candidate_1
+    ) &
+      !is.na(best_candidate_1))
+
+  return(dataframe_no_other)
+}

inst/scripts/plot.R

@@ -13,6 +13,7 @@ source(file = "R/get_gnps.R")
 source(file = "R/get_params.R")
 source(file = "R/log_debug.R")
 source(file = "R/make_confident.R")
+source(file = "R/make_other.R")
 source(file = "R/no_other.R")
 source(file = "R/parse_cli_params.R")
 source(file = "R/parse_yaml_params.R")
@@ -213,17 +214,21 @@ log_debug(x = "keeping peaks similarities with score above", PEAK_SIMILARITY)
 df_new_with_cor <- df_new_with_cor_pre_taxo |>
   dplyr::filter(comparison_score >= PEAK_SIMILARITY)
 
-#' TODO limit to 5 as for hierarchy
 log_debug(x = "plotting histograms")
+#' TODO harmonize 'others' among minor and major
 df_histogram_ready <- df_new_with_cor_pre_taxo |>
+  make_other() |>
   prepare_plot_2()
 df_histogram_outside_ready <- df_peaks_samples_min |>
+  make_other() |>
   prepare_plot_2()
 
 df_histogram_ready_conf <- df_new_with_cor_pre_taxo |>
+  make_other() |>
   no_other() |>
   prepare_plot_2()
 df_histogram_outside_ready_conf <- df_peaks_samples_min |>
+  make_other() |>
   no_other() |>
   prepare_plot_2()
 
@@ -292,10 +297,6 @@ plotly::plot_ly(
   plotly::layout(colorway = sunburst_colors)
 
 
-#' TODO minor peaks
-#' TODO change MS for counts
-
-
 #' Work in progress
 #' Add some metadata per peak
 df_meta <- df_new_with_cor_pre_taxo |>