Data sets used in the empirical comparison study presented in "Prediction approaches for partly missing multi-omics covariate data: A literature review and an empirical comparison study"

These data sets are a subset of the pre-processed versions of the multi-omics datasets originally used in the empirical comparison study presented in Hornung &amp; Wright (2019). In the latter paper, survival was considered as the outcome variable, given by the data.frame "targetvar" (first column "status": status indicator, second column "time": survival times). The remaining objects in each Rda file are "clin", "cnv", "mirna", "mutation", and "rna", which contain clinical data, copy number variation data, miRNA data, mutation data, and RNA data, respectively. Hornung et al. (2023) used the subset of the data sets presented here, where the outcome variable was the presence vs. absence of the TP53 mutation ("1" vs. "0"). This information is provided by the "TP53" column of the mutation data. Note that while predicting this outcome is not contextually meaningful, TP53 mutations have been found to be associated with poor clinical outcomes in cancer patients (Wang &amp; Sun, 2017). In this context, TP53 can be used as a surrogate for a phenotypic outcome. These data sets are intended to test machine learning or statistical methods and may not be useful for biological analysis. <br> References: <br> R. Hornung, F. Ludwigs, J. Hagenberg, and A.-L. Boulesteix. Prediction approaches for partly missing multi-omics covariate data: A literature review and an empirical comparison study. arXiv, arXiv:2302.03991, 2023. <br> R. Hornung, and M. N. Wright. Block Forests: random forests for blocks of clinical and omics covariate data. BMC Bioinformatics, 20:358, 2019. <br> X. Wang, and Q. Sun. TP53 mutations, expression and interaction networks in human cancers. Oncotarget, 8(1):624-643, 2017. <br>