Final figs edits

R/plot_table.R

@@ -35,8 +35,8 @@ plot_table <- function(
     data_integration = "data integration",
     env_collinearity = "environmental variable collinearity",
     env_process = "environmental variable processing",
-    bias = "addressing data biases",
-    study_region = "defining model training extent",
+    bias = "address data biases",
+    study_region = "define model training extent",
     backg_sample = "background data sampling",
     data_partitioning = "occurrence data partitioning",
     mod_fit = "model fitting",
@@ -47,9 +47,9 @@ plot_table <- function(
     mod_multispecies = "multispecies modeling",
     mod_mechanistic = "mechanistic modeling",
     pred_general = "model prediction",
-    pred_extrapolation = "model extrapolation tools/plots",
-    pred_inspect = "model behavior inspection/plots",
-    post_processing = "using predictions as analysis inputs",
+    pred_extrapolation = "model extrapolation tools",
+    pred_inspect = "model behavior inspection",
+    post_processing = "post-processing",
     gui = "graphical user interface",
     metadata = "metadata tools"
   )
@@ -75,19 +75,18 @@ plot_table <- function(
 
   d_n <- rev(unique(dd$category))
   dd$category <- factor(dd$category, levels = d_n)
+  dd <- dd |> dplyr::filter(value == TRUE)
 
-  ggplot(dd, aes(y = category, x = name, fill = value)) +
-    geom_tile() +
-    scale_fill_discrete(
-      guide = "none",
-      type = c("white", "gray30")
-    ) +
+  ggplot(dd, aes(y = category, x = name, fill = category)) +
+    geom_tile(color = "white", fill = "gray20") +
     coord_equal() +
     theme_bw() +
+    scale_x_discrete(position = "top") +
     theme(
-      axis.text.x = element_text(color = pkg_cols, angle = 90, size = 12),
-      axis.text.y = element_text(size = 12),
-      axis.title = element_blank()
-    ) +
-    scale_x_discrete(position = "top")
+      axis.text.x.top =  element_text(color = pkg_cols, angle = 90, size = 14,
+                                      hjust = 0, vjust = 0.5, face = "bold"),
+      axis.text.y = element_text(size = 14),
+      axis.title = element_blank(),
+      legend.position = "none"
+    )
 }

inst/scripts/ms_figs.R

@@ -1,48 +1,37 @@
 library(dplyr)
 
 d <- prep_table(where = "online")
-d$env_process <- ifelse(d$env_collinearity == TRUE, TRUE, d$env_process)
-d$study_region <- ifelse(d$backg_sample == TRUE, TRUE, d$study_region)
-d$pred_inspect <- ifelse(d$pred_extrapolation == TRUE, TRUE, d$pred_inspect)
-d$mod_combine <- d$mod_ensemble
-d$mod_combine <- ifelse(d$mod_stack == TRUE, TRUE, d$mod_combine)
-d <- select(d, -data_integration, -env_collinearity, -backg_sample,
-            -pred_extrapolation, -mod_multispecies, -mod_mechanistic,
-            -mod_stack, -mod_ensemble, -pred_general, -mod_fit) |>
-  filter(name != "rgbif", name != "ibis.iSDM", name != "dismo")
-
-pal <- c("#1b9e77", "#d95f02", "#7570b3", "#e7298a", "#66a61e", "#e6ab02",
-         "#a6761d", "#666666")
-
-pdf("fig3.pdf", height = 7, width = 16)
-f2a <- plot_dendrogram(d, k = 8, cex = 0.7, diff_method = "binary",
-                       k_colors = pal, horiz = TRUE, main = "")
-f2b <- plot_dendrogram(d, k = 7, cex = 1, diff_method = "binary",
-                       type = "phylogenic", repel = TRUE, k_colors = pal)
-gridExtra::grid.arrange(f2a, f2b, nrow = 1, widths = c(10, 6))
-dev.off() #nolint
-
-clust <- plot_dendrogram(d, k = 8, cex = 0.7, diff_method = "binary",
+
+d <- filter(d, name != "rgbif",
+            name != "ibis.iSDM",
+            name != "dismo") |>
+  select(-mod_mechanistic,
+         -mod_multispecies)
+
+pal <- RColorBrewer::brewer.pal(n = 5, name = "Dark2")
+
+pdf("fig3.pdf", height = 5, width = 9.7)
+plot_dendrogram(d, k = 5, cex = 0.7, diff_method = "binary",
+                k_colors = pal, horiz = TRUE, main = "")
+dev.off()
+
+clust <- plot_dendrogram(d, k = 5, cex = 1, diff_method = "binary",
                          k_colors = pal, horiz = TRUE, main = "",
                          return_clust = TRUE)
 pkg_order <- rev(clust$labels[clust$order])
 
-png("fig3.png", height = 480, width = 1100)
-gridExtra::grid.arrange(f2a, f2b, nrow = 1, widths = c(10, 6))
-dev.off() #nolint
-
 pkg_cols <- pkg_order
 names(pkg_cols) <- pkg_order
-pkg_cols[1:4] <- pal[8]
-pkg_cols[5] <- pal[7]
-pkg_cols[6:7] <- pal[6]
-pkg_cols[8:27] <- pal[5]
-pkg_cols[28] <- pal[4]
-pkg_cols[29:31] <- pal[3]
-pkg_cols[32:34] <- pal[2]
-pkg_cols[35] <- pal[1]
-
-g <- plot_table(d,
-                pkg_order = pkg_order,
-                remove_empty_cats = TRUE)
-ggplot2::ggsave("inst/img/pkgs.png", g, height = 10, width = 10)
+
+pkg_cols[1:21] <- pal[5]
+pkg_cols[22:26] <- pal[4]
+pkg_cols[27:30] <- pal[3]
+pkg_cols[31] <- pal[2]
+pkg_cols[32:33] <- pal[1]
+
+pdf("fig2.pdf", height = 7, width = 16)
+plot_table(d,
+           pkg_order = pkg_order,
+           pkg_cols = pkg_cols,
+           remove_empty_cats = TRUE)
+dev.off()