Bayesian pathway selection

We propose a Bayesian pathway selection method that allows the selection of pathways (sets of genes) directly related to a continuous response variable under a non-parametric hierarchical model framework. The fact that sets of genes effectively explain more the response variable than individual genes was the driving force behind this research. We utilize the stochastic search variable selection and kernel machine method to select effective pathways after adjusting clinical covariates effects. The selection of pathways simultaneously works compared to other methods, where pathways are analyzed separately. We show that the proposed model can successfully detect effective pathways associated with outcomes through simulation studies and real data application.