plot.eNetXplorer: generates plots from eNetXplorer object

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

generates plots from eNetXplorer object

Description

This function is a wrapper for a variety of plots, namely:

summary: model performance across alpha (to assess the relative performance among different member models in the elastic net family, as well as in relation to permutation null models);

lambdaVsQF: given alpha, quality function across lambda (to examine the selection of the optimal penalty parameter);

measuredVsOOB: (for gaussian and categorical models) given alpha, response vs out-of-bag predictions across instances (to assess individual instances, examine outliers, etc);

contingency: (for categorical models) given alpha, response vs out-of-bag predictions across classes;

featureCaterpillar: given alpha, caterpillar plot of feature statistics compared to permutation null models (with statistical significance annotations for individual features);

featureHeatmap: heatmap of feature statistics across alpha (including statistical significance annotations for individual features);

KaplanMeier: (for Cox regression models) given alpha, Kaplan-Meier plot of survival probability as a function of time (where the cohort is partitioned in two or more groups based on predicted risk); and

survROC: (for Cox regression models) given alpha, time-dependent ROC plot(s) based on predicted risk at the specified timepoints of interest.

Usage

## S3 method for class 'eNetXplorer'
plot(x, plot.type=c("summary","lambdaVsQF","measuredVsOOB","contingency",
"featureCaterpillar","featureHeatmap","KaplanMeier","survROC"), alpha.index, 
stat=c("freq","coef"), ...)

Arguments

x	eNetXplorer object.
plot.type	Type of plot to be produced. Available plots are "summary", "lambdaVsQF", "measuredVsOOB" (gaussian and categorical models only), "contingency" (categorical models only), "featureCaterpillar", "featureHeatmap", "KaplanMeier" (Cox models only) and "survROC" (Cox models only).
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. Used for plot types "featureHeatmap" and "featureCaterpillar", ignored otherwise.
...	Additional plotting parameters.

Author(s)

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

See Also

eNetXplorer, plotSummary, plotLambdaVsQF, plotMeasuredVsOOB, plotContingency,

plotFeatureCaterpillar, plotFeatureHeatmap, plotKaplanMeier, plotSurvROC

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="summary")
plot(x=fit,plot.type="lambdaVsQF",alpha.index=2)
plot(x=fit,plot.type="measuredVsOOB",alpha.index=c(1,3,5))
plot(x=fit,plot.type="featureCaterpillar",stat="coef")
plot(x=fit,plot.type="featureHeatmap",stat="freq")

juliancandia/eNetXplorer documentation built on May 11, 2023, 1:59 a.m.