separate2GroupsCox: Separate data in High and Low risk groups (based on Cox...

In glmSparseNet: Network Centrality Metrics for Elastic-Net Regularized Models

Description

Draws multiple kaplan meyer survival curves (or just 1) and calculates logrank test

Usage

separate2GroupsCox(
  chosen.btas,
  xdata,
  ydata,
  probs = c(0.5, 0.5),
  no.plot = FALSE,
  plot.title = "SurvivalCurves",
  xlim = NULL,
  ylim = NULL,
  expand.yzero = FALSE,
  legend.outside = FALSE,
  ...
)

Arguments

chosen.btas	list of testing coefficients to calculate prognostic indexes, for example “list(Age = some_vector)“
xdata	n x m matrix with n observations and m variables
ydata	Survival object
probs	How to separate high and low risk patients 50%-50% is the default, but for top and bottom 40% -> c(.4,.6)
no.plot	Only calculate p-value and do not generate survival curve plot
plot.title	Name of file if
xlim	Optional argument to limit the x-axis view
ylim	Optional argument to limit the y-axis view
expand.yzero	expand to y = 0
legend.outside	If TRUE legend will be outside plot, otherwise inside
...	additional parameters to survminer::ggsurvplot

Value

object with logrank test and kaplan-meier survival plot

A list with plot, p-value and kaplan-meier object. The plot was drawn from survminer::ggsurvplot with only the palette, data and fit arguments being defined and keeping all other defaults that can be customized as additional parameters to this function.

See Also

survminer::ggsurvplot

Examples

data('cancer', package = 'survival')
xdata <- ovarian[,c('age', 'resid.ds')]
ydata <- data.frame(time = ovarian$futime, status = ovarian$fustat)
separate2GroupsCox(c(age = 1, 0), xdata, ydata)
separate2GroupsCox(c(age = 1, 0.5), xdata, ydata)
separate2GroupsCox(c(age = 1), c(1,0,1,0,1,0),
                   data.frame(time = runif(6), status = rbinom(6, 1, .5)))
separate2GroupsCox(list(aa = c(age = 1, 0.5),
                        bb = c(age = 0, 1.5)), xdata, ydata)

glmSparseNet documentation built on April 14, 2021, 6 p.m.