R package TreatmentSelection

An R package containing a suite of descriptive and inferential methods designed to evaluate individual treatment selection markers and to compare candidate markers.

functions included are:

• trtsel_measures for estimating performance measures for a pre-specified marker based treatment rule
• trtsel for creating trtsel objects
• plot for plotting risk curves and more
• evaluate for evaluating marker performance
• calibrate for assessing model calibration
• compare to compare two trtsel objects.

An older stable version of the package is available from CRAN. However, this version does not allow for multivariate models or time-to-event outcomes and will be phased out soon.

install.packages("TreatmentSelection")

An updated version of the package with extended capabilities is currently under development. To install this version directly from github, type:

if (!require("devtools")) install.packages("devtools")
devtools::install_github("mdbrown/TreatmentSelection")

A tutorial highlighting the capabilities of the development version of the package (from github) is here: Tutorial

References

Janes H, Brown MD, Huang Y, Pepe MS. 2014. An Approach to Evaluating and Comparing Biomarkers for Patient Treatment Selection. International Journal of Biostatistics