deprecate Taxa names -> Features (#91)

* deprecate Taxa names -> Features * deprecated Taxa names and ggplot and tidyr args * .data deprecated in tidyselect * deprecated aes_string * deprecated ggplot and tidyr args and code consistency * fix .get_abundance_data * restore aes_args * restore .data * fix aes_string * add disable note * documentation update * \U2012 mismatch in code and docs * disable notes and code documentation mismatch * fix disable notes * fix disable notes * documentation update * fix list --------- Co-authored-by: Tuomas Borman <[email protected]>
microbiome · Jan 29, 2024 · adc8e9e · adc8e9e
1 parent 51449d6
commit adc8e9e
Show file tree

Hide file tree

Showing 17 changed files with 103 additions and 72 deletions.
diff --git a/NAMESPACE b/NAMESPACE
@@ -5,6 +5,7 @@ export(plotCCA)
 export(plotColGraph)
 export(plotColTile)
 export(plotDMNFit)
+export(plotFeaturePrevalence)
 export(plotPrevalence)
 export(plotPrevalentAbundance)
 export(plotRDA)
@@ -23,6 +24,7 @@ exportMethods(plotColGraph)
 exportMethods(plotColTile)
 exportMethods(plotColTree)
 exportMethods(plotDMNFit)
+exportMethods(plotFeaturePrevalence)
 exportMethods(plotPrevalence)
 exportMethods(plotPrevalentAbundance)
 exportMethods(plotRDA)
@@ -74,7 +76,7 @@ importFrom(dplyr,ungroup)
 importFrom(ggnewscale,new_scale)
 importFrom(ggnewscale,new_scale_colour)
 importFrom(ggnewscale,new_scale_fill)
-importFrom(ggplot2,aes_string)
+importFrom(ggplot2,aes)
 importFrom(ggplot2,coord_flip)
 importFrom(ggplot2,element_blank)
 importFrom(ggplot2,element_text)

diff --git a/R/plotAbundance.R b/R/plotAbundance.R
@@ -112,7 +112,7 @@
 #' # Getting top taxa on a Phylum level
 #' se <- transformAssay(se, method="relabundance")
 #' se_phylum <- agglomerateByRank(se, rank ="Phylum", onRankOnly=TRUE)
-#' top_taxa <- getTopTaxa(se_phylum,top = 5, assay.type = "relabundance")
+#' top_taxa <- getTopFeatures(se_phylum,top = 5, assay.type = "relabundance")
 #' 
 #' # Renaming the "Phylum" rank to keep only top taxa and the rest to "Other"
 #' phylum_renamed <- lapply(rowData(se)$Phylum,
@@ -274,7 +274,7 @@ MELT_VALUES <- "Value"
     data <- data %>%
         as.data.frame() %>%
         mutate(rank = factor(rowData(x)[,rank], unique(rowData(x)[,rank]))) %>%
-        pivot_longer(cols = !.data$rank,
+        pivot_longer(cols = !rank,
                     names_to = MELT_NAME,
                     values_to = MELT_VALUES) %>%
         mutate(!!sym(MELT_NAME) := factor(!!sym(MELT_NAME), unique(!!sym(MELT_NAME)))) %>%
@@ -292,7 +292,7 @@ MELT_VALUES <- "Value"
         lvl <- lvl[order(lvl)]
     } else if(order_rank_by %in% c("abund","revabund")){
         o <- data %>% 
-            select(!.data$X) %>% 
+            select(!.data$X) %>%
             group_by(.data$colour_by) %>% 
             summarize(sum = sum(.data$Y))
         decreasing <- ifelse(order_rank_by == "abund",TRUE,FALSE)
@@ -304,7 +304,7 @@ MELT_VALUES <- "Value"
     }
     data$colour_by <- factor(data$colour_by, lvl)
     data <- data[order(data$colour_by),]
-    #
+
     data
 }
 
@@ -424,7 +424,7 @@ MELT_VALUES <- "Value"
                         point_alpha = 1,
                         point_size = 2){
     # start plotting
-    plot_out <- ggplot(object, aes_string(x="X", y="Y")) +
+    plot_out <- ggplot(object, aes(x=.data[["X"]], y=.data[["Y"]])) +
         xlab(xlab) +
         ylab(ylab)
     # either bar or point plot
@@ -471,7 +471,7 @@ MELT_VALUES <- "Value"
     plot_out
 }
 
-#' @importFrom ggplot2 ggplot aes_string labs geom_point geom_raster
+#' @importFrom ggplot2 ggplot aes labs geom_point geom_raster
 .feature_plotter <- function(feature_data,
                             name,
                             xlab,
@@ -485,7 +485,7 @@ MELT_VALUES <- "Value"
         feature_data$Y <- ""
         colour_by <- unique(feature_data$feature_name)
     }
-    feature_plot_out <- ggplot(feature_data, aes_string(x="X", y="Y")) +
+    feature_plot_out <- ggplot(feature_data, aes(x=.data[["X"]], y=.data[["Y"]])) +
         labs(x = xlab, y = name)
     if(length(unique(feature_data$Y)) == 1L){
         feature_out <- .get_bar_args(colour_by,

diff --git a/R/plotAbundanceDensity.R b/R/plotAbundanceDensity.R
@@ -214,7 +214,7 @@ setMethod("plotAbundanceDensity", signature = c(object = "SummarizedExperiment")
     # Gets the assay
     mat <- assay(object, assay.type, withDimnames = TRUE)
     # Gets the most abundant taxa
-    top_taxa <- getTopTaxa(object, top = n, assay.type = assay.type)
+    top_taxa <- getTopFeatures(object, top = n, assay.type = assay.type)
     # Subsets abundance table  by taking taxa of highest abundance
     mat <- mat[top_taxa, , drop=FALSE]
     # enable conversion to data.frame for non-matrix assays, e.g. sparseMatrices
@@ -282,7 +282,7 @@ setMethod("plotAbundanceDensity", signature = c(object = "SummarizedExperiment")
                              scales_free = TRUE,
                              angle_x_text = TRUE){
     # start plotting
-    plot_out <- ggplot(density_data, aes_string(x="X")) +
+    plot_out <- ggplot(density_data, aes(x=.data[["X"]])) +
         xlab(xlab) +
         ylab(ylab)
     # Layout can be "density", "jitter", or "point"

diff --git a/R/plotCCA.R b/R/plotCCA.R
@@ -500,7 +500,6 @@ setMethod("plotRDA", signature = c(object = "matrix"),
         parse.labels = TRUE, vec.text = TRUE, repel.labels = TRUE, add.ellipse = TRUE,
         position = NULL, nudge_x = NULL, nudge_y = NULL, direction = "both",
         max.overlaps = 10, check_overlap = FALSE, ...){
-
     # Get the scatter plot
     plot <- plot_data[["plot"]]
     # Add ellipse
@@ -517,7 +516,7 @@ setMethod("plotRDA", signature = c(object = "matrix"),
                              aes(x = .data[[xvar]], y = .data[[yvar]],
                                  color = .data[[colour_var]], fill = after_scale(color)),
                              geom = "polygon", alpha = ellipse.alpha,
-                             size = ellipse.linewidth, linetype = ellipse.linetype)
+                             linewidth = ellipse.linewidth, linetype = ellipse.linetype)
         } else if ( add.ellipse %in% c("color", "colour") ){
             plot <- plot +
                 stat_ellipse(data = data,

diff --git a/R/plotDMN.R b/R/plotDMN.R
@@ -46,7 +46,7 @@ setGeneric("plotDMNFit", signature = "x",
 
 #' @rdname plotDMN
 #' @importFrom DirichletMultinomial mixture
-#' @importFrom ggplot2 ggplot aes_string geom_point geom_line theme_bw labs
+#' @importFrom ggplot2 ggplot aes geom_point geom_line theme_bw labs
 #' @export
 setMethod("plotDMNFit", signature = c(x = "SummarizedExperiment"),
     function(x, name = "DMN", type = c("laplace","AIC","BIC")){
@@ -62,7 +62,7 @@ setMethod("plotDMNFit", signature = c(x = "SummarizedExperiment"),
         #
         k <- vapply(dmn, function(d){ncol(mixture(d))}, numeric(1))
         fit <- vapply(dmn, fit_FUN, numeric(1))
-        ggplot(data.frame(k = k, fit = fit), aes_string(x = k, y = fit)) +
+        ggplot(data.frame(k = k, fit = fit), aes(x = k, y = fit)) +
           geom_point() +
           geom_line() +
           theme_bw() +

diff --git a/R/plotGraph.R b/R/plotGraph.R
@@ -644,10 +644,11 @@ setMethod("plotRowGraph",
 
 #' @importFrom tidyr drop_na
 .add_graph_labels <- function(plot_out, show_label){
+    label <- NULL # disable note: no global binding for variable
     if(show_label){
-        label_data <- plot_out$data %>% drop_na(.data$label)
+        label_data <- plot_out$data %>% drop_na(label)
         plot_out <- plot_out +
-            geom_node_label(mapping = aes_string(label = "label"), 
+            geom_node_label(mapping = aes(label = .data[["label"]]), 
                             data = label_data,
                             repel = TRUE,
                             max.overlaps = 100)

diff --git a/R/plotPrevalence.R b/R/plotPrevalence.R
@@ -1,9 +1,9 @@
 #' Plot prevalence information
 #' 
-#' \code{plotPrevalence} and \code{plotTaxaPrevalence} visualize prevalence 
+#' \code{plotPrevalence} and \code{plotFeaturePrevalence} visualize prevalence 
 #' information.
 #' 
-#' Whereas \code{plotPrevalence} produces a line plot, \code{plotTaxaPrevalence}
+#' Whereas \code{plotPrevalence} produces a line plot, \code{plotFeaturePrevalence}
 #' returns a heatmap. 
 #' 
 #' @param x a
@@ -100,14 +100,14 @@
 #' plotPrevalence(GlobalPatterns, rank = "Phylum") + scale_x_log10()
 #' 
 #' # plotting prevalence per taxa for different detection thresholds as heatmap
-#' plotTaxaPrevalence(GlobalPatterns, rank = "Phylum")
+#' plotFeaturePrevalence(GlobalPatterns, rank = "Phylum")
 #' 
 #' # by default a continuous scale is used for different detection levels, 
 #' # but this can be adjusted
-#' plotTaxaPrevalence(GlobalPatterns, rank = "Phylum",
+#' plotFeaturePrevalence(GlobalPatterns, rank = "Phylum",
 #'                    detections = c(0, 0.001, 0.01, 0.1, 0.2))
 #'                    
-#' # point layout for plotTaxaPrevalence can be used to visualize by additional
+#' # point layout for plotFeaturePrevalence can be used to visualize by additional
 #' # information
 #' plotPrevalentAbundance(GlobalPatterns, rank = "Family",
 #'                        colour_by = "Phylum") +
@@ -337,16 +337,17 @@ setMethod("plotPrevalentAbundance", signature = c(x = "SummarizedExperiment"),
 }
 
 ################################################################################
-# plotTaxaPrevalence
+# plotFeaturePrevalence
 
 #' @rdname plotPrevalence
+#' @aliases plotTaxaPrevalence
 #' @export
-setGeneric("plotTaxaPrevalence", signature = c("x"),
-           function(x, ...) standardGeneric("plotTaxaPrevalence"))
+setGeneric("plotFeaturePrevalence", signature = c("x"),
+           function(x, ...) standardGeneric("plotFeaturePrevalence"))
 
 #' @rdname plotPrevalence
 #' @export
-setMethod("plotTaxaPrevalence", signature = c(x = "SummarizedExperiment"),
+setMethod("plotFeaturePrevalence", signature = c(x = "SummarizedExperiment"),
           function(x,
                    rank = taxonomyRanks(x)[1L],
                    assay.type = assay_name, assay_name = "counts",
@@ -444,7 +445,7 @@ setMethod("plotTaxaPrevalence", signature = c(x = "SummarizedExperiment"),
     lvls <- rownames(ans)[order(rowSums(ans))]
     ans$ID <- rownames(mat)[rowSums(f) != 0]
     ans <- ans %>%
-        pivot_longer(!.data$ID, 
+        pivot_longer(!ID, 
                      names_to = "detection",
                      values_to = "prevalence")
     colnames(ans) <- c("Y","X","colour_by")
@@ -474,7 +475,7 @@ setMethod("plotTaxaPrevalence", signature = c(x = "SummarizedExperiment"),
                                 line_alpha = 1,
                                 line_type = NULL,
                                 line_size = 1){
-    plot_out <- ggplot(plot_data, aes_string(x = "X", y = "Y")) +
+    plot_out <- ggplot(plot_data, aes(x = .data[["X"]], y = .data[["Y"]])) +
         labs(x = xlab, y = ylab)
     if(layout == "line"){
         point_args <- .get_point_args(colour_by = colour_by, shape_by = NULL,
@@ -485,7 +486,7 @@ setMethod("plotTaxaPrevalence", signature = c(x = "SummarizedExperiment"),
                                     size_by = NULL,
                                     alpha = line_alpha,
                                     linetype = line_type,
-                                    size = line_size)
+                                    linewidth = line_size)
         point_args$args$mapping$group <- sym("colour_by")
         line_args$args$mapping$group <- sym("colour_by")
         plot_out <- plot_out +
@@ -542,3 +543,20 @@ setMethod("plotTaxaPrevalence", signature = c(x = "SummarizedExperiment"),
                            angle_x_text = FALSE)
     plot_out
 }
+
+#' @rdname plotPrevalence
+#' @aliases plotFeaturePrevalence
+#' @export
+setGeneric("plotTaxaPrevalence", signature = c("x"),
+           function(x, ...) 
+             standardGeneric("plotTaxaPrevalence"))
+
+#' @rdname plotPrevalence
+#' @aliases plotFeaturePrevalence
+#' @export
+setMethod("plotTaxaPrevalence", signature = c(x = "ANY"),
+          function(x, ...){
+            .Deprecated(old ="plotTaxaPrevalence", new = "plotFeaturePrevalence", msg = "The 'plotTaxaPrevalence' function is deprecated. Use 'plotFeaturePrevalence' instead.")
+            plotFeaturePrevalence(x, ...)
+          }
+)
diff --git a/R/plotSeries.R b/R/plotSeries.R
@@ -63,7 +63,7 @@
 #' # Plots 2 most abundant taxa, which are colored by their family
 #' plotSeries(object,
 #'            x = "DAY_ORDER",
-#'            y = getTopTaxa(object, 2),
+#'            y = getTopFeatures(object, 2),
 #'            colour_by = "Family")
 #' 
 #' # Counts relative abundances
@@ -82,7 +82,7 @@
 #' # In addition to 'colour_by' and 'linetype_by', 'size_by' can also be used to group taxa.
 #' plotSeries(object,
 #'            x = "DAY_ORDER", 
-#'            y = getTopTaxa(object, 5), 
+#'            y = getTopFeatures(object, 5), 
 #'            colour_by = "Family",
 #'            size_by = "Phylum",
 #'            assay.type = "counts")
@@ -304,7 +304,7 @@ setMethod("plotSeries", signature = c(object = "SummarizedExperiment"),
     }
     # Creates a "draft" of a plot
     plot_out <- ggplot(plot_data,
-                       aes_string(x = "X", y = "Y")) +
+                       aes(x = .data[["X"]], y = .data[["Y"]])) +
         labs(x = xlab, y = ylab)
     # if sd column is present add a ribbon
     if(!is.null(plot_data$sd)){
@@ -313,13 +313,13 @@ setMethod("plotSeries", signature = c(object = "SummarizedExperiment"),
         plot_out <- plot_out +
             do.call(geom_ribbon, ribbon_args$args)
     }
-    # Fetches arguments fpr geom_line
+    # Fetches arguments for geom_line
     line_args <- .get_line_args(colour_by = colour_by,
                                 linetype_by = linetype_by,
                                 size_by = size_by,
                                 alpha = line_alpha,
                                 linetype = line_type,
-                                size = line_width)
+                                linewidth = line_width)
     # Adds arguments to the plot
     plot_out <- plot_out +
         do.call(geom_line, line_args$args)
@@ -359,7 +359,7 @@ setMethod("plotSeries", signature = c(object = "SummarizedExperiment"),
         guide_args$linetype <- guide_legend(title = linetype_by)
     }
     if (!is.null(size_by)) {
-        guide_args$size <- guide_legend(title = size_by)
+        guide_args$linewidth <- guide_legend(title = size_by)
     }
     if (length(guide_args)) {
         plot_out <- plot_out + do.call(guides, guide_args)

diff --git a/R/plotTree.R b/R/plotTree.R
@@ -112,7 +112,7 @@
 #'   log(rowData(altExp(GlobalPatterns,"Genus"))$mean)
 #' rowData(altExp(GlobalPatterns,"Genus"))$detected <-
 #'    rowData(altExp(GlobalPatterns,"Genus"))$detected / 100
-#' top_genus <- getTopTaxa(altExp(GlobalPatterns,"Genus"),
+#' top_genus <- getTopFeatures(altExp(GlobalPatterns,"Genus"),
 #'                         method="mean",
 #'                         top=100L,
 #'                         assay.type="counts")
@@ -149,7 +149,7 @@
 #' # please note that the original tree of GlobalPatterns is dropped by
 #' # unsplitByRanks
 #' altExps(GlobalPatterns) <- splitByRanks(GlobalPatterns)
-#' top_phyla <- getTopTaxa(altExp(GlobalPatterns,"Phylum"),
+#' top_phyla <- getTopFeatures(altExp(GlobalPatterns,"Phylum"),
 #'                         method="mean",
 #'                         top=10L,
 #'                         assay.type="counts")

diff --git a/R/treeData.R b/R/treeData.R
@@ -118,7 +118,7 @@ setMethod("rowTreeData", signature = c(x = "TreeSummarizedExperiment"),
 DEFAULT_TREE_DATA_COLS <- c("parent","node","branch.length","label")
 .clean_tree_data <- function(tree_data){
     tree_data %>%
-        select(DEFAULT_TREE_DATA_COLS)
+        select(all_of(DEFAULT_TREE_DATA_COLS))
 }
 
 #' @rdname treeData