plot.hrunbiasedDiagnostic: Cox model diagnostic plots using gene random signatures

In adricaba/hrunbiased: Survival analysis for gene signatures

Description

Plot function for hrunbiasedDiagnostic objects

Usage

## S3 method for class 'hrunbiasedDiagnostic'
plot(x, diagnostic.plot, id.size = 1,
   show.stat = TRUE, legend.out = TRUE, perms.to.show = 1:10,
   seq.alpha = seq(0,.5, length.out=50), seq.alpha.pow = c(0.01,0.05),
   power.whplots = NULL, col1 = "gray", col2 = "orange",
   pointsq = FALSE, ...)

Arguments

x	Object of class hrunbiasedDensityRS, hrunbiasedPermutations, hrunbiasedSimulations or hrunbiasedPCpower
diagnostic.plot	At least one of "density.rs", "density.rs.size", "geneMean.geneSign" if x is of class hrunbiasedDensityRS, "perm.violin", "perm.GSvsEvents", "perms.corr.GS" if x is of class hrunbiasedPermutations, "simulations" if x is of class hrunbiasedSimulations, and "positive.cont.power", "bias.power", "corNCsig.power" if x is of class hrunbiasedPCpower, "PCAcorrelation", if x is of class hrunbiasedPCAcorrelation
id.size	For class hrunbiasedDensityRS related plots. Show random signature distribution for signature size sigSize[id.size]
show.stat	For "density.rs" plots. If TRUE, hazard ratio statistic is shown, if FALSE, log hazard ratio is shown
legend.out	If TRUE, an automatic legend is shown in the plot
perms.to.show	For "permutations" diagnostic plots. Number of instances to show (keep it low for interpretability)
seq.alpha	For "positive.cont.power" diagnostic plots. Sequence of rejecting levels to be used to plot power curves
seq.alpha.pow	For "corNCsig.power" or "bias.power" diagnostic plots. Sequence of rejecting levels to be used to summarize the power using several negative control levels
power.whplots	For class hrunbiasedPCpower related plots. Show power plots for contaminated positive control signature defined by prop.pos.cont[power.whplots]
col1	For "simulations" diagnostic plot. Color for non-adjusted random signature distributions
col2	For "simulations" diagnostic plot. Color for GS-adjusted random signature distributions
pointsq	If TRUE, population logHR for target signature is shown
...	Arguments passed to or from other methods to the low level.

Details

Following details section in hrunbiasedDiagnostic, module (i) contains "density.rs": hazard ratio t-statistic density plot obtained from random signatures; "density.rs.size": density plot for hazard ratio t-statistic as function of signature size; "geneMean.geneSign": average log HR genewise in a signature vs log HR in the signature. Module (ii) contains "perm.violin": hazard ratio violin plots obtained from random signatures for several instances with a shuffle in the time-to-even outcome; "perm.GSvsEvents": global signature boxplots that distinguish between event and not event; "perms.corr.GS": relationship between GS event and not event average difference and observed average lHR. Module (iii) contains "positive.cont.power": power curves using positive control random signatures; "bias.power": two y-axis plot showing average values for the hazard ratio t-statistic and power given several charactarizations of negative controls; "corNCsig.power": two y-axis plot showing correlation to positive controls and power of several charactarizations of negative controls; Module (iv) contains "simulations": random signatures hazard ratio distributions for simulated time-to-event data.

Value

plot of object of class hrunbiasedDiagnostic.

Author(s)

Adria Caballe Mestres

References

Caballe Mestres A, Berenguer Llergo A and Stephan-Otto Attolini C. Adjusting for systematic technical biases in risk assessment of gene signatures in transcriptomic cancer cohorts. bioRxiv (2018).

See Also

hrunbiasedDiagnostic

Examples

data(out.brca)
plot(out.brca[["metabric"]], diagnostic.plot = "density.rs", main = "metabric",
     xlab = "lhr.stat", ylab="", lwd=3, legend.out = FALSE, id.size = 3 )

adricaba/hrunbiased documentation built on May 24, 2019, 7:48 a.m.