By Stuart G. Baker, 2015
Introduction
A predictive marker is a baseline variable in a randomized trial that is used to determine subgroups in which the effect of treatment is greater than average. The software splits the data into training and test samples. The model fits a benefit function to the training sample and applies the fitted benefit function to marker data from each test sample participant to obtain a benefit score for each test sample participant. Results are summarized in a subpopulation treatment effect pattern plot that indicates the estimated treatment effect at various cutpoints of the benefit score. This implementation applies to survival times with a censoring indicator.
Requirement: Mathematica Version 10 or later.
Set-Up
| copy | all files into some folder called "FOLDER" |
|---|---|
| start | a new Mathematica session |
| type | SetDirectory["FOLDER"] |
| type | << survfit.m |
To run on hypothetical data,
type TrialFitSurv[datasim]
To try on your own data,
type TrialFitSurv[ dataset, options]
Options
| Option | Default | Explanation |
|---|---|---|
| NewFitQ | True | New fitting or use stored result of previous fit |
| Time | .1 | Time at which survival is computed |
| FractionSplit | 0.5 | Fraction split into test sample |
| ModelSet | "Basic" | Risk difference with 1, 2, 5 markers |
| MinimumGroupSize | 100 | Minimum sample size of training-test sample |
| Num Cut | 8 | Number of cutpoints in test sample |
| ShowProgQ | False | Show progress of fitting algorithm |
| ShowTabQ | False | Show data and results tables |
| MaxBoot | 20 | Number of bootstraps for confidence interval |
| PlotFormat | "quantiles" | Format of horizontal axis |
dataset ={x0,x1,(t0,c0),(t1,c1),xname,datasetname}
| x0 | n x g matrix of baseline variables for randomization group 0 |
|---|---|
| x1 | n x g matrix of baseline variables for randomization group 1 |
| c0 | a length n list of censoring indicators (0 or 1=cens) for randomization group 0 |
| c1 | a length n list of censoring indicators (0 or 1=cens) for randomization group 1 |
| t0 | a length n list of binary outcomes (0 or 1) for randomization group 0 |
| t1 | a length n list of binary outcomes (0 or 1) for randomization group 1 |
| xname | a length g list of names of baseline variables |
| datasetname | name of dataset |
File Contents
Download All (ZIP, 46 KB)
| File name | Description |
|---|---|
| survfit.m | Calls all files |
| survfitinputcheck | Checks input |
| survfitmodel.m | Calls appropriate model |
| survfitplot.m | Plot STEPP |
| survfitsurv.m | Key function for creating novel outcome measure, 0, 1, or fraction |
| survfitsplit.m | Split data into training and test samples |
| survfitlogit.m | Fit logistic regression |
| survfitbasic.m | Calls test sample analysis with bootstrap confidence intervals |
| survfitbasictest.m | Basic computations for test sample analysis |
| survfitboot.m | Bootstrap computations |
| survfitbootci.m | Boostrap confidence interval |
| survfitrd.m | Risk difference |
| survfitrdtrain.m | Risk difference analysis for training sample |
| survfitrdtest.m | Risk difference: test sample |
| survfitsim.m | Generate simulated data |
Disclaimer
This code is provided "as is", without warranty of any kind, express or implied, including but not limited to the warranties of merchantability, fitness for a particular purpose and non-infringement. In no event shall the NCI or the individual developers be liable for any claim, damages or other liability of any kind. Use of this code by recipient is at recipient's own risk. NCI makes no representations that the use of the code will not infringe any patent or proprietary rights of third parties.
Last updated: January 05, 2016