Produces a heatmap as a grid grob.
Usage
plot_heatmap(
y,
cluster_samples = FALSE,
cluster_features = TRUE,
sample_labels = NULL,
feature_labels = NULL,
baseline = NULL,
baseline_label = "row\nmean",
scale_label = "difference from\nrow mean",
n = Inf,
baseline_to = NULL,
scale_to = NULL,
show_baseline = TRUE,
show_tree = TRUE
)Arguments
- y
A matrix of expression levels, such as a transformed counts matrix as produced by
varistran::vst.- cluster_samples
Should samples (columns) be clustered?
- cluster_features
Should features (rows) be clustered?
- sample_labels
Names for each sample. If not given and y has column names, these will be used instead.
- feature_labels
Names for each feature. If not given and y has row names, these will be used instead.
- baseline
Baseline level for each row, to be subtracted when drawing the heatmap colors. If omitted, the row mean will be used. Specify
baseline=0to not subtract anything and not show a baseline bar graph.- baseline_label
Text description of what the baseline is.
- scale_label
Text description of what the heatmap colors represent (after baseline is subtracted).
- n
Show only this many rows. Rows are selected in order of greatest span of expression level.
- baseline_to
If provided, the scale for the row means or baseline will include this value or these values. Use this if there is some meaningful "zero" for your data.
- scale_to
If provided, the heatmap color scale will include this value. Use this to provide consistency of scales between heatmaps (note values larger than scale_to will cause the scale to be extended).
- show_baseline
Show baseline barplot?
- show_tree
Show dendrogram tree(s)? These dendrograms arguably over-interpret the data without adding much of values, so it may be better to hide them.
Value
A grid grob. print()-ing this value will cause it to be displayed.
Additionally $info$row_order will contain row ordering and $info$col_order will contain column ordering.
Details
This heatmap differs from other heatmaps in R in the method of clustering used:
1. The distances used are cosine distances (i.e. the magnitude of log fold changes is not important, only the pattern).
2. hclust() is used to produce a clustering, as normal.
3. Branches in the hierarchical clustering are flipped to minimize sharp changes between neighbours, using the seriation package's OLO (Optimal Leaf Ordering) method.
