Bayesian methods for expression-based integration

摘要: We propose methods to integrate data across several genomic platforms using a hierarchical Bayesian analysis framework that incorporates the biological relationships among identify genes whose expression is related clinical outcomes in cancer. This integrated approach combines information all platforms, leading increased statistical power finding these predictive genes, and further provides mechanistic about manner which gene affects outcome. demonstrate advantages of shrinkage estimation used by this through simulation, finally, we apply our method Glioblastoma Multiforme dataset potentially associated with patients’ survival. find 12 positive prognostic markers nine 13 negative genes.