surv_fitness: Survival fitness function using the Restricted Mean Survival...

In harpomaxx/GSgalgoR: An Evolutionary Framework for the Identification and Study of Prognostic Gene Expression Signatures in Cancer

Description

Survival fitness function using the Restricted Mean Survival Time (RMST) of each group as proposed by Dehbi & Royston et al. (2017).

Usage

surv_fitness(OS, clustclass, period)

Arguments

OS	a survival object with survival data of the patients evaluated
clustclass	a numeric vector with the group label for each patient
period	a number representing the period of time to evaluate in the RMST calculation

Value

The function computes the Harmonic mean of the differences between Restricted Mean Survival Time (RMST) of consecutive survival curves multiplied by the number of comparisons.

Author(s)

Martin E Guerrero-Gimenez, [e-mail]

References

Dehbi Hakim-Moulay, Royston Patrick, Hackshaw Allan. Life expectancy difference and life expectancy ratio: two measures of treatment effects in randomized trials with non-proportional hazards BMJ 2017; 357 :j2250 https://www.bmj.com/content/357/bmj.j2250

Examples

# load example dataset
library(breastCancerTRANSBIG)
library(Biobase)
data(transbig)
Train <- transbig
rm(transbig)

clinical <- pData(Train)
OS <- survival::Surv(time = clinical$t.rfs, event = clinical$e.rfs)

surv_fitness(OS, clustclass = clinical$grade, period = 3650)

harpomaxx/GSgalgoR documentation built on Oct. 25, 2020, 3:47 p.m.