CosmonetTesting: Make predictions on data
In cosmonet-package/COSMONET: Screening-network Cox methods for survival analysis

This function performes the model validation and prediction using the testing set D. It generates the Kaplan-Meier curve resulting from the log-rank test between high and low-risk group and the distribution plot of prognostic index PI^{D} computed using the signature genes and the optimal cutoff PI^{*,T} obtained from the training set T.

1	CosmonetTesting(x, y, screenVars, beta, opt.cutoff)

`x`	input testing matrix `nxp`. Each row is an observation vector.
`y`	response variable, `y` should be a two-column data frame with columns named time and status. The latter is a binary variable, with 1 indicating event, and 0 indicating right censored. The rownames indicate the sample names ordered as the samples in the input testing matrix.
`screenVars`	screened variables obtained from BMD- or DAD-, or BMD+DAD-screening.
`beta`	regression coefficients estimated on the training set.
`opt.cutoff`	optimal cutoff selected adaptively on the training set `T`.

The following objects are returned:

`df`	data frame composed by prognostic index `PI^{D}`, sample, time and status.
`p.value`	resulting from the log-rank test (the significance level is `p`-value < 0.05).