RLasso-Cox

"RLasso-Cox" addresses the challenge of accurately predicting cancer patient risk using high-dimensional gene expression data. The tool enhances the Cox proportional hazard model with the Lasso for variable selection, introducing a Reweighted Lasso-Cox (RLasso-Cox) model. This new model incorporates gene interaction information, considering the topological importance of genes in gene interaction networks. By prioritizing topologically significant genes with stable expression changes, RLasso-Cox demonstrates improved prognostic accuracy and robustness compared to existing methods, particularly on independent datasets.

Topic

Molecular interactions, pathways and networks;Biomarkers;Oncology;Gene expression;Microarray experiment

Detail

  • Operation: Gene prediction;Regression analysis;Feature selection

  • Software interface: Command-line user interface

  • Language: R

  • License: Artistic License 2.0

  • Cost: Free

  • Version name: 1.11.0

  • Credit: The National Natural Science Foundation of China, the Innovation Team Project of Heilongjiang Institute of Technology, National Social Science Foundation of China, the Heilongjiang Social science planning project.

  • Input: -

  • Output: -

  • Contact: Wei Liu freelw@qq.com

  • Collection: -

  • Maturity: Stable

Publications

  • Integration of gene interaction information into a reweighted Lasso-Cox model for accurate survival prediction.
  • Wang W and Liu W. Integration of gene interaction information into a reweighted Lasso-Cox model for accurate survival prediction. Integration of gene interaction information into a reweighted Lasso-Cox model for accurate survival prediction. 2021; 36:5405-5414. doi: 10.1093/bioinformatics/btaa1046
  • https://doi.org/10.1093/BIOINFORMATICS/BTAA1046
  • PMID: 33325490
  • PMC: -

Download and documentation

< Back to DB search