How to use the rbsurv Package
暂无分享,去创建一个
The rbsurv package is designed to select survival-associated genes, based on a likelihood function. It utilizes the partial likelihood of the Cox model which has been the basis for many of the existing methods. Our algorithm is simple and straight-forward, but its functions such as the generation of multiple gene models and the incorporation of significant risk factors are practical. For robustness, this package also selects survivalassociated genes by separating train and validation sets of samples because such a crossvalidation technique is essential in predictive modeling for data with large variability. It employs forward selection, the limitation of which is mitigated by generating a series of gene models and selecting an optimal model. Furthermore, iterative runs after putting aside the previously selected genes can discover the masked genes that may be missed by forward selection (see Cho et al. for details). The rbsurv package employs libraries survival and Biobase.
[1] Jaewoo Kang,et al. Robust Likelihood-Based Survival Modeling with Microarray Data , 2009 .
[2] S. Horvath,et al. Gene Expression Profiling of Gliomas Strongly Predicts Survival , 2004, Cancer Research.