Skip to contents

Heatmap of RVD sequence variants across strains and TALE groups

Source: R/classification.R
talomes_heatmap.Rd

Draws a talome overview: one column per TALE group, one row per strain, each cell showing which RVD sequence variant that strain carries in that group. A strain's talome is its whole complement of TALEs, so the plot shows at a glance which groups each strain has and where strains carry different variants of the same TALE.

Within a group, the cell colour gives each variant's rank by the number of strains that carry it: the most common variant is the palest and rarer ones are darker. The # after each group label counts its distinct variants. A white cell means the strain has no member in that group. A cell can hold several colours side by side when a strain carries more than one variant in the same group. Dendrograms order strains and groups by the similarity of their variant profiles.

Usage

talomes_heatmap(
  tale_annotation,
  group_col = "group",
  strain_col,
  rvd_col,
  trunc_tales_col = NULL,
  extra_col = NULL,
  x_lab = "TALE Group",
  y_lab = "Strain",
  title = "RVD sequences variants",
  plot_type = "all",
  colors = .tantale_colours$variant_ranks,
  margins = NULL,
  sep_width = 5,
  sep_color = "white",
  inner_sep_color = "white",
  save_path = NULL
)

Arguments

tale_annotation

A data frame with one row per TALE and at least a group, a strain and an RVD-sequence column. Or a tales object carrying group and strain columns, with one value per array (for instance a grouped tales to which a strain column was added): the RVD sequences are then computed with tales_rvd_strings, and arrays without repeats are left out.

group_col

Name of the tale_annotation column holding TALE groups, drawn as columns. The default, "group", is the column tales_group_kmedoids and tales_group_hclust add.

strain_col

Name of the column holding strain names, drawn as rows.

rvd_col

Name of the column holding RVD sequences (e.g. from tales_rvd_strings). Not needed when tale_annotation is a tales object.

trunc_tales_col

Optional name of a logical column marking truncated TALEs; those are labelled "T" in their cell (with plot_type = "all").

extra_col

Optional name of a column with further information about each strain (e.g. origin), drawn as a side bar on the right.

x_lab, y_lab, title

Axis names and plot title.

plot_type

Either "all" to draw every allele, or "single" to draw one representative allele per group.

colors

Colours from the most common variant to the rarest. They are interpolated over the ranks present, so the most common variant always gets the first colour and the rarest the last. The default runs from pale to dark wine, so a rare variant stands out.

margins

Space for the row dendrogram, column dendrogram, row names and column names, in that order, counted in heatmap cells. NULL (default) sizes the dendrograms to a quarter of the heatmap's width and height, between 1.5 and 5 cells. With plot_type = "all", it gives the names and the title the room their text takes, whatever the size of the device, and the cells share the rest; with plot_type = "single", the names get 3 cells, and the column dendrogram also holds the title and gets at least 3.5 cells.

sep_width

Width of the separator between adjacent cells.

sep_color

Colour of the separator between adjacent cells.

inner_sep_color

Colour of the separator between variants within one cell.

save_path

Optional file path; the image format follows the file extension. If NULL (default), the heatmap is drawn on the current device.

Value

NULL, invisibly. Called for the side effect of drawing the heatmap (or, if save_path is given, writing it to a file).

See also

Other TALE plots: plot.tales(), plot.tales_msa(), plot_target_preds()

Examples

ann <- data.frame(
  group = c("G1", "G1", "G1", "G2", "G2"),
  strain = c("S1", "S2", "S3", "S1", "S2"),
  rvdseq = c("NI-HD-NG", "NI-HD-NG", "NN-HD-NG",
            "HD-NI-NG-NG", "HD-NI-NG-NG")
)
talomes_heatmap(ann, group_col = "group", strain_col = "strain",
                rvd_col = "rvdseq")


# From a tales object: one group and one strain per array
x <- tales_from_telltales(system.file("extdata", "tellTaleExampleOutput",
                                      package = "tantale"))
x$group <- c(ROI_00001 = 1, ROI_00002 = 1, ROI_00003 = 2, ROI_00004 = 2)[x$array_id]
x$strain <- c(ROI_00001 = "S1", ROI_00002 = "S2", ROI_00003 = "S1",
              ROI_00004 = "S2")[x$array_id]
talomes_heatmap(x, strain_col = "strain")