BANFF: An R Package for BAyesian Network Feature Finder

Feature selection on high-dimensional networks plays an important role in understanding of biological mechanisms and disease pathologies. It has a broad range of applications. Recently, a Bayesian nonparametric mixture model (Zhao, Kang, and Yu 2014) has been successfully applied for selecting gene and gene sub-networks. We extend this method to a unified approach for feature selection on general high-dimensional networks; and we develop a powerful R package, the Bayesian network feature finder (BANFF), providing a full package of posterior inference, model comparison, and graphical illustration of model fitting. In BANFF, we develop a parallel computing algorithm for the Markov chain Monte Carlo (MCMC) based posterior inference and an ExpectationMaximization (EM) based algorithm for posterior approximation, both of which greatly reduce the computational time for model inference. In this work, we provide detailed instruction on how to use the R functions in BANFF along with several tutorial examples on analysis of simulated datasets and real datasets. Particularly, we demonstrate the use of BANFF on selecting features from a protein-protein interaction network and perform brain image segmentations.