Supplementary files - Boulesteix et al. (2015)
IPF-LASSO: integrative L1-penalized regression with penalty factors for prediction based on multi-omics data
Contact: Anne-Laure Boulesteix
The R-codes to reproduce the analyses presented in the paper are available here.
- The simulation study can be reproduced by running the script 'simulation.script.r' (by Anne-Laure Boulesteix). This script loads the functions defined in the file 'simulation.fun.r', creates two directories 'results_uncorrelated' and 'results_correlated' and saves the results there.
- The figures displaying the results of the simulation can be generated by running the script 'figures.script.r'. This script loads the functions defined in the file 'figures.fun.r' and the results stored in the directories 'results_uncorrelated' and 'results_correlated'.
- The real data studies (AML data, breast cancer with survival outcome, breast cancer with binary outcome) can be reproduced by running the scripts 'realdata1.script_AML.r', 'realdata2.script_breastcancer_survival.r' and 'realdata3.script_breastcancer_logistic.r' (by Riccardo De Bin), respectively. These scripts also generate the figures.
- reproducibility (17 KByte)