fixes related to nr_children column

R/AnalysisConfiguration.R

@@ -397,35 +397,24 @@ hierarchy_counts <- function(pdata, config){
 #' config <- bb$config
 #' data <- bb$data
 #' res <- hierarchy_counts_sample(data, config, nr_children = 1)
-#' res()
 #' x <- res("long")
-#' res("plot")
 #' # filters on peptide level
 #' res <- hierarchy_counts_sample(data, config, nr_children = 2)
-#' res()
 #' x2 <- res("long")
-#' res("plot")
 #' # filters on protein level based on peptide count
 #' bb <- prolfqua::sim_lfq_data_protein_config()
 #' res <- hierarchy_counts_sample(bb$data, bb$config, nr_children = 2)
 #' x1 <- res()
-#' res("long")
-#' res("plot")
 #' res <- hierarchy_counts_sample(bb$data, bb$config, nr_children = 1)
 #' x2 <- res()
-#' res("long")
-#' res("plot")
 #' x1$nr_children <- 2
 #' x2$nr_children <- 1
 #' xl <- dplyr::bind_rows(x1, x2)
 #'
-#'
-#'
-#'
 #' xl$nr_children |> table()
 #' nudgeval <-  -mean(xl$protein_Id) * 0.05
-#' ggplot(xl, aes(x = sampleName, y = protein_Id, fill = as.character(nr_children)) ) +
-#'  geom_bar(stat = "identity", position = position_dodge())
+#' ggplot2::ggplot(xl, ggplot2::aes(x = sampleName, y = protein_Id, fill = as.character(nr_children)) ) +
+#'  ggplot2::geom_bar(stat = "identity", position = ggplot2::position_dodge())
 #'
 hierarchy_counts_sample <- function(
     pdata,

R/ContrastsPlotter.R

@@ -76,7 +76,7 @@ ContrastsPlotter <- R6::R6Class(
     fcthresh = 1,
     #' @field avg.abundance name of column containing avg abundance values.
     avg.abundance = character(),
-    #' @field avg.abundance name of column containing avg abundance values.
+    #' @field protein_annot protein annotation
     protein_annot = NULL,
     #' @description
     #' create Crontrast_Plotter
@@ -90,6 +90,7 @@ ContrastsPlotter <- R6::R6Class(
     #' @param diff fold change (difference) diff column
     #' @param contrast contrast column
     #' @param avg.abundance name of column with average abundance
+    #' @param protein_annot add protein annotation (optional)
     initialize = function(contrastDF,
                           subject_Id,
                           volcano = list(list(score = "FDR", thresh = 0.1)),

R/ProteinAnnotation.R

@@ -23,31 +23,39 @@ ProteinAnnotation <-
                 pID = character(),
                 #' @field description name of column containing descriptions
                 description = "description",
-                #' @field ids vector with columns containing addition IDs
-                ids = character(),
+                #' @field cleaned_ids vector with columns containing addition IDs
+                cleaned_ids = character(),
                 #' @field nr_children name of columns with the number of peptides
                 nr_children = character(),
                 #' @description initialize
                 #' @param lfqdata data frame from \code{\link{setup_analysis}}
                 #' @param row_annot data frame with row annotation. Must have columns matching \code{config$table$hierarchy_keys_depth()}
                 #' @param description name of column with description
-                #' @param ids names of columns with additional ID's
+                #' @param ids names of columns with cleaned Ids
                 #' @param nr_peptides additional peptides
+                #' @param nr_children column with the number of children
                 initialize = function(lfqdata,
                                       row_annot,
                                       description = NULL,
                                       ids = NULL,
                                       nr_children = "nr_peptides"){
                   self$pID = lfqdata$config$table$hierarchy_keys_depth()[1]
                   self$nr_children = nr_children
-                  if (!missing(row_annot)) {
+                  if ( !is.null(ids)) {self$cleaned_ids = ids} else {self$cleaned_ids = self$pID}
+                  if ( !is.null(description)) {self$description = description} else {self$description = self$pID}
+                  if ( !missing(row_annot)) {
                     stopifnot(self$pID %in% colnames(row_annot))
                     row_annot <- dplyr::filter(row_annot, !!sym(self$pID) %in% lfqdata$data[[self$pID]] )
                     self$row_annot <- row_annot
                   } else {
                     self$row_annot <- distinct(select(lfqdata$data, self$pID))
                   }
+
+                  stopifnot(self$cleaned_ids %in% colnames(self$row_annot))
+                  stopifnot(self$description %in% colnames(self$row_annot))
+
                   if (!self$nr_children %in% colnames(row_annot) ) {
+                    warning("no nr_children column specified, computing using nr_obs_experiment function")
                     self$row_annot <- inner_join(
                       self$row_annot,
                       nr_obs_experiment(lfqdata$data, lfqdata$config, name_nr_child = self$nr_children),

R/tidyMS_R6_TransitionCorrelations.R

@@ -645,7 +645,7 @@ nr_B_in_A <- function(pdata, config , merge = TRUE){
 #' @keywords internal
 #' @family summary
 #' @examples
-#' debug(nr_B_in_A_per_sample)
+#'
 #' bb <- prolfqua::sim_lfq_data_peptide_config()
 #' nr_B_in_A_per_sample(bb$data, bb$config, nested =FALSE)
 #' bb <- prolfqua::sim_lfq_data_protein_config()

R/tidyMS_aggregation.R

@@ -752,7 +752,7 @@ aggregate_intensity_topN <- function(pdata , config, .func, N = 3){
     workIntensity = newcol,
     hierarchy = config$table$hierarchy[seq_len(config$table$hierarchyDepth)])
 
-  new_child_name <- paste0("nr_", config$table$hierarchy_keys_depth() )
+  new_child_name <- paste(c("nr", config$table$hierarchy_keys_depth()), collapse = "_" )
   res_nr_children <- nr_obs_sample(pdata, config, new_child = new_child_name)
   sumTopInt <- inner_join(
     sumTopInt, res_nr_children,