Annotation observations

library(ggalign)
#> Loading required package: ggplot2
#> ========================================
#> ggalign version 1.2.0.9000
#> 
#> If you use it in published research, please cite: 
#> Peng, Y.; Jiang, S.; Song, Y.; et al. ggalign: Bridging the Grammar of Graphics and Biological Multilayered Complexity. Advanced Science. 2025. doi:10.1002/advs.202507799
#> ========================================

set.seed(123)
# Prepare heatmap matrix
mat <- matrix(
    rnorm(100, mean = 0, sd = 2),
    nrow = 10, ncol = 10,
    dimnames = list(paste0("G", 1:10), paste0("S", 1:10))
)

By default, when no manual observations were selected, ggmark() will select all observations and split them based on the groups defined in the layout.

set.seed(123)
ggheatmap(mat) +
    anno_right() +
    align_kmeans(3L) +
    ggmark(mark_line()) +
    geom_boxplot(aes(.names, value)) +
    theme(plot.margin = margin(l = 0.1, t = 0.1, unit = "npc"))
#> → heatmap built with `geom_tile()`

You can also annotate a specific subset of observations. For example, to annotate only rows 1–3 and 6–10:

set.seed(123)
ggheatmap(mat) +
    anno_right() +
    align_kmeans(3L) +
    # Select and annotate rows 1–3 and 6–10 only
    ggmark(mark_line(1:3, 6:10)) +
    geom_boxplot(aes(.names, value)) +
    theme(plot.margin = margin(l = 0.1, t = 0.1, unit = "npc"))
#> → heatmap built with `geom_tile()`

For more details, please visit https://yunuuuu.github.io/ggalign-book/annotate-observations.html