annHeatmap2 {Heatplus} | R Documentation |
This function plots a data matrix as intensity heatmap, with optional dendrograms, annotation panels and clustering for both rows and columns. This is the actual working function called by numerous wrappers.
annHeatmap2( x, dendrogram, annotation, cluster, labels, scale = c("row", "col", "none"), breaks = 256, col = g2r.colors, legend = FALSE )
x |
the numerical matrix to be shown as heatmap |
dendrogram |
a list that controls how row- and column diagrams are determined and displayed |
annotation |
a list that controls the data and the way it is shown in row- and column annotation panels |
cluster |
a list that controls how many clusters are chosen, and how these clusters are labelled and colored |
labels |
a list that controls the row- and column labels, as well as their size and placement |
scale |
a character string indicating how the matrix |
breaks |
specifies the interval breaks for displaying the data in
|
col |
specifies a palette of colors for the heatmap intensities; either
a vector of explicit color definitions (one less than breaks) or a palette
function. See |
legend |
whether and where to draw a legend for the colors/intervals in
the heatmap. If |
Arguments scale
, breaks
, col
and legend
control
different aspects of the whole plot directly as described. Arguments
dendrogram
, annotation
, cluster
and labels
control aspects that may differ for the rows and columns of the central
heatmap and have a special structure: each is a named list with different
entries controling e.g. the look of a dendrogram, the data for annotation
etc. Additionally, they can contain two extra entries called simply
Row
and Col
; these are again named lists that can contain all
the same entries as the parent list. Entries specified directly in the list
apply to both rows and columns; entries specified as part of Row
or
Col
override these defaults for the rows or columns only.
Recognized parameters for argument dendrogram
:
the clustering function for generating the dendrogram;
defaults to hclust
for rows and columns
a function
that returns the pairwise distances between samples/features as an object of
class dist
; defaults to dist
for rows and columns
a string that controls the display of the dendrogram:
yes
means use the dendrogram to re-order the rows/columns and display
the dendrogram; hidden
means re-rorder, but do not display; no
means do not use the dendrogram at all.
the line width of the branches of the dendrogram; defaults to 3.
an override
argument that allows to pass in a dendrogram directly, bypassing the
clustfun
and distfun
mechanism; defaults to NULL
(i.e.
is not used)
Recognized entries for argument annotation
:
a
data frame containing the annotation data; defaults to NULL
, i.e. no
annotation is displayed
a list of fine-tuning parameters
that is passed directly to picketPlot
; defaults to an empty list,
i.e. the default settings in picketPlot
logical value
indicating whether the annotation data
needs to be pre-processed via
convAnnData
or not; defaults to TRUE
logical
value indicating whether to include all levels of factor variables in
data
, or whether to drop the reference level (i.e. the first level).
Defaults to TRUE
Recognized entries for argument cluster
:
the
height at which to cut through the dendrogram to define groups of similar
features/samples; defaults to NULL
, i.e. no cutting
labels for the clusters; defaults to NULL
, i.e. no
labels
colors for the different clusters; the colors are used
for coloring both the sub-trees of the dendrogram and the corresponding area
in the annotation plot (if there is one). This is either a vector of colors,
or a palette function that takes a number and returns a vector of colors of
the specified length; defaults to BrewerClusterCol
an
override argument that directly specifies group memberships for the
features/samples, completely bypassing the whole dendrogram
and
cuth
mechanism. This probably only works for
dendrogram$status="no"
.
Recognized entries for argument labels
:
size of
the text for the labels; defaults to NULL
, i.e. use a hard-coded
default guess
amount of space available for the labels between the central heatmap and the dendrogram, expressed as lines of text; defaults to 3.
side at which to draw the labels, coded as integer between 1 and 4 in the usual way (1 = below the plot, continuing clockwise). A common default for rows and columns does not make sense: rows only work with 2 and 4, columns only with 1 and 3. Defaults try to make use of empty space, depending on the presence of a dendrogram.
labels for
rows and columns; defaults to NULL
, i.e. using the row- and column
names of x
. Note that these labels are applied after
re-sorting rows and columns as per dendrogram, so these have to be already
sorted accordingly. If you want to change the labels before
re-sorting, it is is easier to re-set the row- and/or column names of
x
.
An object of class annHeatmap
. Use plot
to display it
graphically.
heatmapLayout
, niceBreaks
,
breakColors
, g2r.colors
, BrewerClusterCol
require(Biobase) data(sample.ExpressionSet) ex1 = sample.ExpressionSet[51:85,] map1 = annHeatmap2(exprs(ex1), ann=list(Col=list(data=pData(ex1))), cluster=list(Col=list(cuth=3000))) plot(map1)