The spike-and-slab lasso Cox model for survival prediction and associated genes detection

• DOI: 10.1093/bioinformatics/btx300
• Publication Date: 2017-05-03T19:11:11Z
• Authors: Zaixiang Tang; Yueping Shen; Xinyan Zhang; Nengjun Yi
• Keywords: Lasso; R package

Large-scale molecular profiling data have offered extraordinary opportunities to improve survival prediction of cancers and other diseases detect disease associated genes. However, there are considerable challenges in analyzing large-scale data.We propose new Bayesian hierarchical Cox proportional hazards models, called the spike-and-slab lasso Cox, for predicting outcomes detecting We also develop an efficient algorithm fit proposed models by incorporating Expectation-Maximization steps into extremely fast cyclic coordinate descent algorithm. The performance method is assessed via extensive simulations compared with regression. demonstrate procedure on two cancer datasets censored thousands features. Our analyses suggest that can generate powerful prognostic genes.The methods been implemented a freely available R package BhGLM ( http://www.ssg.uab.edu/bhglm/ )[e-mail] at Bioinformatics online.