plotFeatureHeatmap: generates heatmap plot of feature statistics

In eNetXplorer: Quantitative Exploration of Elastic Net Families for Generalized Linear Models

Description

This function generates a heatmap plot of feature statistics across alpha, which includes statistical significance annotations for individual features. By default, features are selected (and ordered top-down) by statistical significance based on a given value of alpha; options are provided to customize feature selection and display.

Usage

plotFeatureHeatmap(x, alpha.index=NULL, stat=c("freq","coef"), feature.all=FALSE, 
feature.pval.thres=NULL, feature.set=NULL, feature.top.n=25, signif.code=TRUE, 
xlab=NULL, ylab=NULL, main=NULL, col.main="black", cex.main=0.95, line=1, 
col=NULL, breaks=NULL, scale="none", Rowv=FALSE, Colv=FALSE, na.color=NULL, 
cexRow=NULL, srtRow=0, cexCol=0.75, srtCol=45, margins=c(5, 5), key=TRUE, 
key.title=NA, dendogram="none", trace="none", notecol.freq="black", 
notecol.coef="white", notecex=1, subtitle1=NULL, col.subtitle1="black", 
line.subtitle1=-1, cex.subtitle1=0.65, subtitle2=NULL, col.subtitle2="darkgray", 
line.subtitle2=-2, cex.subtitle2=0.55, ...)

Arguments

x	eNetXplorer object.
alpha.index	Integer indices to select alpha values. Default is 1:length(alpha)
stat	Feature statistic: "freq" for nonzero frequency, "coef" for mean nonzero coefficient.
feature.all	(Feature selection argument 1) Logical to show all features. Default is FALSE.
feature.pval.thres	(Feature selection argument 2) P-value threshold to select features. Default is NULL.
feature.set	(Feature selection argument 3) Character vector of feature names to display. Default is NULL.
feature.top.n	(Feature selection argument 4) Number of top features (ordered by p-value) to display. Default is 25.
signif.code	Logical to display statistical significance annotations. Default is TRUE.
xlab	Label for x axis.
ylab	Label for y axis.
main	Custom title.
col.main	Title color.
cex.main	Title size.
line	Title location.
col	Heatmap color vector. Length must be one less than number of breaks.
breaks	Color breaks vector. Default number of breaks is 10.
scale	Logical to scale the data for heatmap in either the row or column direction. Default is "none".
Rowv	Logical to reorder rows by hierarchical clustering. Default is FALSE.
Colv	Logical to reorder columns by hierarchical clustering. Default is FALSE.
na.color	Color to use for missing values.
cexRow	Size of row labels.
srtRow	Angle of row labels, in degrees from horizontal.
cexCol	Size of column labels.
srtCol	Angle of column labels, in degrees from horizontal.
margins	Numeric vector of length 2 containing the margins for column and row names, respectively.
key	Logical to display key. Default is TRUE.
key.title	Main title of the color key.
dendogram	To draw dendograms. Default is "none".
trace	To display trace lines. Default is "none".
notecol.freq	Color of statistical significance annotations for feature frequency heatmaps.
notecol.coef	Color of statistical significance annotations for feature coefficient heatmaps.
notecex	Size of significance annotations.
subtitle1	Custom subtitle 1.
col.subtitle1	Color of subtitle 1.
line.subtitle1	Position of subtitle 1.
cex.subtitle1	Size of subtitle 1.
subtitle2	Custom subtitle 2.
col.subtitle2	Color of subtitle 2.
line.subtitle2	Position of subtitle 2.
cex.subtitle2	Size of subtitle 2.
...	Additional plotting parameters.

Details

Feature selection criteria are hierarchical based on arguments 1 through 4 (see argument description above), with argument 1 at the top of the hierarchy. E.g. if feature.all is explicitly set to TRUE, it will take precedence over any other feature selection argument. By default, the top 25 features are displayed, ordered top-down by significance based on the given value of alpha.

Author(s)

Julian Candia and John S. Tsang
Maintainer: Julian Candia [e-mail]

See Also

eNetXplorer, plot

Examples

data(QuickStartEx)
fit = eNetXplorer(x=QuickStartEx$predictor,y=QuickStartEx$response,
family="gaussian",n_run=20,n_perm_null=10,seed=111)
plot(x=fit,plot.type="featureHeatmap",stat="freq")
plotFeatureHeatmap(x=fit,alpha.index=3,stat="freq",main="custom title")

eNetXplorer documentation built on Aug. 30, 2020, 1:06 a.m.