Comprehensive group splited cluster plot on dimension reduction with proportion
Source:R/visualization.R
plotGroupClusterDimRed.Rd
This function produces combined plot on group level (e.g. dataset, other metadata variable like biological conditions). Scatter plot of dimension reduction with cluster labeled is generated per group. Furthermore, a stacked barplot of cluster proportion within each group is also combined with the subplot of each group.
Arguments
- object
A liger object with dimension reduction, grouping variable and cluster assignment in
cellMeta(object)
.- useGroup
Variable name of the group division in metadata. Default
"dataset"
.- useCluster
Name of variable in
cellMeta(object)
. DefaultNULL
uses default cluster.- useDimRed
Name of the variable storing dimensionality reduction result in
cellMeta(object)
. DefaultNULL
use default dimRed.- combinePlot
Whether to return combined plot. Default
TRUE
. IfFALSE
, will return a list containing only the scatter plots.- droplevels
Logical, whether to perform
droplevels()
on the selected grouping variable. DefaultTRUE
will not show groups that are listed as categories but do not indeed have any cells.- relHeightMainLegend
Relative heights of the main combination panel and the legend at the bottom. Must be a numeric vector of 2 numbers. Default
c(5, 1)
.- relHeightDRBar
Relative heights of the scatter plot and the barplot within each subpanel. Must be a numeric vector of 2 numbers. Default
c(10, 1)
.- mainNRow, mainNCol
Arrangement of the main plotting region, for number of rows and columns. Default
NULL
will be automatically handled byplot_grid
.- legendNRow
Arrangement of the legend, number of rows. Default
1
.- ...
Additional graphic setting arguments passed to
plotDimRed
.
Value
ggplot object when only one feature (e.g. cluster variable, gene, factor) is set. List object when multiple of those are specified.
See also
Please refer to plotDimRed
,
.ggScatter
, .ggplotLigerTheme
for additional
graphic setting