Layout Customization

The package offers a suite of align_* functions designed to give you precise control over plot layout. These functions enable you to reorder the observations or partition the observations into multiple groups.

Currently, there are four key align_* functions available for layout customization:

align_group: Group and align plots based on categorical factors.
align_order: Reorder layout observations based on statistical weights or allows for manual reordering based on user-defined ordering index.
align_kmeans: Group observations by k-means clustering results.
align_dendro: Align plots according to hierarchical clustering or dendrograms.

library(ggalign)

set.seed(123)
small_mat <- matrix(rnorm(81), nrow = 9)
rownames(small_mat) <- paste0("row", seq_len(nrow(small_mat)))
colnames(small_mat) <- paste0("column", seq_len(ncol(small_mat)))

`align_group`

The align_group() function allows you to group rows/columns into separate panels. It doesn’t add any plot area.

ggheatmap(small_mat) +
    anno_top() +
    align_group(sample(letters[1:4], ncol(small_mat), replace = TRUE))
#> → heatmap built with `geom_tile()`

By default, the facet strip text is removed. You can override this behavior with theme(strip.text = element_text()). Since align_group() does not create a new plot, the panel title can only be added to the heatmap plot.

ggheatmap(small_mat) +
    theme(strip.text = element_text()) +
    anno_left() +
    align_group(sample(letters[1:4], nrow(small_mat), replace = TRUE))
#> → heatmap built with `geom_tile()`

`align_order`

The align_order() function order the rows/columns based on the summary weights, Like align_group(), it doesn’t add a plot area.

Here, we order the rows based on the means.

ggheatmap(small_mat) +
    anno_left() +
    align_order(rowMeans)
#> → heatmap built with `geom_tile()`

In addition, we can provide the ordering integer index directly in the order argument.

my_order <- sample(nrow(small_mat))
print(rownames(small_mat)[my_order])
#> [1] "row3" "row1" "row7" "row6" "row2" "row8" "row9" "row5" "row4"
ggheatmap(small_mat) +
    anno_left() +
    align_order(my_order)
#> → heatmap built with `geom_tile()`

We can also provide the ordering character index.

ggheatmap(small_mat) +
    anno_left() +
    align_order(rownames(small_mat)[my_order])
#> → heatmap built with `geom_tile()`

By default, align_order() reorders the rows or columns in ascending order of the summary function’s output (from bottom to top for rows, or from left to right for columns). To reverse this order, you can set reverse = TRUE:

ggheatmap(small_mat) +
    anno_left() +
    align_order(rowMeans, reverse = TRUE)
#> → heatmap built with `geom_tile()`

Some align_* functions accept a data argument. This can be a matrix, a data frame, or even a simple vector. The data argument can also accept a function (purrr-like lambda syntax is supported), which will be applied to the layout data. It is important to note that all align_* functions consider the rows as the observations. It means the NROW(data) must return the same number with the observations in axis used for alignment.

quad_layout()/ggheatmap(): for column annotation, the layout data will be transposed before using (If data is a function, it will be applied with the transposed matrix). This is necessary because column annotation uses heatmap columns as observations, but we need rows.
stack_layout()/ggstack(): the layout data will be used as it is since we place all plots along a single axis.

Even for top and bottom annotations, you can use rowMeans() to calculate the mean value across all columns.

ggheatmap(small_mat) +
    anno_top() +
    align_order(rowMeans)
#> → heatmap built with `geom_tile()`

`align_kmeans`

The align_kmeans() function groups heatmap rows or columns based on k-means clustering. Like the previous functions, it does not add a plot area.

ggheatmap(small_mat) +
    anno_top() +
    align_kmeans(3L)
#> → heatmap built with `geom_tile()`

Note that all align_* functions which define groups must not break the previous established groups. This means the new groups must nest in the old groups, in this way, usually they cannot be used if groups already exist.

ggheatmap(small_mat) +
    anno_top() +
    align_group(sample(letters[1:4], ncol(small_mat), replace = TRUE)) +
    align_kmeans(3L)
#> Error in `align_kmeans()`:
#> ! `align_kmeans()` disrupt the previously established panel groups of
#>   the layout x-axis

ggheatmap(small_mat) +
    anno_top() +
    align_kmeans(3L) +
    align_group(sample(letters[1:4], ncol(small_mat), replace = TRUE))
#> Error in `align_group()`:
#> ! `align_group()` disrupt the previously established panel groups of the
#>   layout x-axis

align_dendro

The align_dendro() function adds a dendrogram to the layout and can also reorder or split the layout based on hierarchical clustering. This is particularly useful for working with heatmap plots.

ggheatmap(small_mat) +
    anno_top() +
    align_dendro()
#> → heatmap built with `geom_tile()`

Hierarchical clustering is performed in two steps: calculate the distance matrix and apply clustering. You can use the distance and method argument to control the dendrogram builind process.

There are two ways to specify distance metric for clustering:

specify distance as a pre-defined option. The valid values are the supported methods in dist() function and coorelation coefficient "pearson", "spearman" and "kendall". The correlation distance is defined as 1 - cor(x, y, method = distance).
a self-defined function which calculates distance from a matrix. The function should only contain one argument. Please note for clustering on columns, the matrix will be transposed automatically.

ggheatmap(small_mat) +
    anno_top() +
    align_dendro(distance = "pearson") +
    patch_titles(top = "pre-defined distance method (1 - pearson)")
#> → heatmap built with `geom_tile()`

ggheatmap(small_mat) +
    anno_top() +
    align_dendro(distance = function(m) dist(m)) +
    patch_titles(top = "a function that calculates distance matrix")
#> → heatmap built with `geom_tile()`

Method to perform hierarchical clustering can be specified by method. Possible methods are those supported in hclust() function. And you can also provide a self-defined function, which accepts the distance object and return a hclust object.

ggheatmap(small_mat) +
    anno_top() +
    align_dendro(method = "ward.D2")
#> → heatmap built with `geom_tile()`

The dendrogram can also be used to cut the columns/rows into groups. You can specify k or h, which work similarly to cutree():

ggheatmap(small_mat) +
    anno_top() +
    align_dendro(k = 3L)
#> → heatmap built with `geom_tile()`

In contrast to align_group(), align_kmeans(), and align_order(), align_dendro() is capable of drawing plot components. So it has a default set_context value of TRUE, meaning it will set the active context of the annotation stack layout. In this way, we can add any ggplot elements to this plot area.

ggheatmap(small_mat) +
    anno_top() +
    align_dendro() +
    geom_point(aes(y = y))
#> → heatmap built with `geom_tile()`

The align_dendro() function creates default node data for the ggplot. See ggplot2 specification in ?align_dendro for details. Additionally, edge data is added to the ggplote::geom_segment() layer directly, used to draw the dendrogram tree. One useful variable in both node and edge data is the branch column, corresponding to the cutree result:

ggheatmap(small_mat) +
    anno_top() +
    align_dendro(aes(color = branch), k = 3) +
    geom_point(aes(color = branch, y = y))
#> → heatmap built with `geom_tile()`

You can reorder the dendrogram based on the mean values of the observations by setting reorder_dendrogram = TRUE.

h1 <- ggheatmap(small_mat) +
    anno_top() +
    align_dendro(aes(color = branch), k = 3, reorder_dendrogram = TRUE) +
    ggtitle("reorder_dendrogram = TRUE")
h2 <- ggheatmap(small_mat) +
    anno_top() +
    align_dendro(aes(color = branch), k = 3) +
    ggtitle("reorder_dendrogram = FALSE")
align_plots(h1, h2)
#> → heatmap built with `geom_tile()`
#> → heatmap built with `geom_tile()`

align_dendro() can also perform clustering between groups, meaning it can be used even if there are existing groups present in the layout, in this way, you cannot specify k or h:

set.seed(3L)
column_groups <- sample(letters[1:3], ncol(small_mat), replace = TRUE)
ggheatmap(small_mat) +
    anno_top() +
    align_group(column_groups) +
    align_dendro(aes(color = branch))
#> → heatmap built with `geom_tile()`

You can reorder the groups by setting reorder_group = TRUE.

ggheatmap(small_mat) +
    anno_top() +
    align_group(column_groups) +
    align_dendro(aes(color = branch), reorder_group = TRUE)
#> → heatmap built with `geom_tile()`

You can merge the sub-tree in each group by settting merge_dendrogram = TRUE.

ggheatmap(small_mat) +
    anno_top() +
    align_group(column_groups) +
    align_dendro(aes(color = branch), merge_dendrogram = TRUE)
#> → heatmap built with `geom_tile()`

You can reorder the dendrogram and merge simutaneously.

ggheatmap(small_mat) +
    anno_top() +
    align_group(column_groups) +
    align_dendro(aes(color = branch),
        reorder_group = TRUE,
        merge_dendrogram = TRUE
    ) +
    anno_bottom() +
    align_dendro(aes(color = branch),
        reorder_group = FALSE,
        merge_dendrogram = TRUE
    )
#> → heatmap built with `geom_tile()`

If you specify k or h, this will always turn off sub-clustering. The same principle applies to align_dendro(), where new groups must be nested within the previously established groups.

ggheatmap(small_mat) +
    anno_top() +
    align_group(column_groups) +
    align_dendro(k = 2L)
#> Error in `align_dendro()`:
#> ! `align_dendro()` disrupt the previously established panel groups of
#>   the layout x-axis

Session information

sessionInfo()
#> R version 4.4.2 (2024-10-31)
#> Platform: x86_64-pc-linux-gnu
#> Running under: Ubuntu 24.04.1 LTS
#> 
#> Matrix products: default
#> BLAS:   /usr/lib/x86_64-linux-gnu/openblas-pthread/libblas.so.3 
#> LAPACK: /usr/lib/x86_64-linux-gnu/openblas-pthread/libopenblasp-r0.3.26.so;  LAPACK version 3.12.0
#> 
#> locale:
#>  [1] LC_CTYPE=en_US.UTF-8       LC_NUMERIC=C              
#>  [3] LC_TIME=en_US.UTF-8        LC_COLLATE=C              
#>  [5] LC_MONETARY=en_US.UTF-8    LC_MESSAGES=en_US.UTF-8   
#>  [7] LC_PAPER=en_US.UTF-8       LC_NAME=C                 
#>  [9] LC_ADDRESS=C               LC_TELEPHONE=C            
#> [11] LC_MEASUREMENT=en_US.UTF-8 LC_IDENTIFICATION=C       
#> 
#> time zone: Etc/UTC
#> tzcode source: system (glibc)
#> 
#> attached base packages:
#> [1] stats     graphics  grDevices utils     datasets  methods   base     
#> 
#> other attached packages:
#> [1] rmarkdown_2.29     ggalign_0.0.5.9000 ggplot2_3.5.1      bookdown_0.41     
#> 
#> loaded via a namespace (and not attached):
#>  [1] gtable_0.3.6       jsonlite_1.8.9     compiler_4.4.2     Rcpp_1.0.13-1     
#>  [5] ggbeeswarm_0.7.2   magick_2.8.5       jquerylib_0.1.4    textshaping_0.4.0 
#>  [9] systemfonts_1.1.0  scales_1.3.0       yaml_2.3.10        fastmap_1.2.0     
#> [13] R6_2.5.1           labeling_0.4.3     knitr_1.49         tibble_3.2.1      
#> [17] maketools_1.3.1    munsell_0.5.1      RColorBrewer_1.1-3 bslib_0.8.0       
#> [21] pillar_1.9.0       rlang_1.1.4        utf8_1.2.4         cachem_1.1.0      
#> [25] xfun_0.49          sass_0.4.9         sys_3.4.3          viridisLite_0.4.2 
#> [29] cli_3.6.3          withr_3.0.2        magrittr_2.0.3     digest_0.6.37     
#> [33] grid_4.4.2         beeswarm_0.4.0     lifecycle_1.0.4    vipor_0.4.7       
#> [37] ggrastr_1.0.2      vctrs_0.6.5        evaluate_1.0.1     glue_1.8.0        
#> [41] farver_2.1.2       ragg_1.3.3         buildtools_1.0.0   fansi_1.0.6       
#> [45] colorspace_2.1-1   tools_4.4.2        pkgconfig_2.0.3    htmltools_0.5.8.1