Bayesian Nonparametric Differential Analysis for Dependent Multigroup Data with Application to Colorectal Cancer DNA Methylation
Chiyu Gu, Veerabhadran Baladandayuthapani, Jeffrey S. Morris, Subharup Guha · arXiv (Cornell University) · 2017
With recent advances in array-based and next-generation sequencing technologies, cancer datasets involve widely varying sizes and scales, measurement variables, and correlation structures. An overarching scientific goal in cancer research is the invention of general statistical techniques that can cleanly sift the signal from the noise in identifying genomic signatures of the disease across a set of experimental or biological conditions. We propose BayesDiff, a nonparametric Bayesian approach based on a novel class of first order mixture models, called the Sticky Poisson-Dirichlet process or multicuisine restaurant franchise. The BayesDiff methodology flexibly utilizes information from all the measurements and adaptively accommodates any serial dependence in the data to perform simultaneous inferences on the variables. The technique is applied to analyze the motivating DNA methylation colorectal cancer dataset, which displays both serial correlations and complex interaction patterns. In simulation studies, we demonstrate the effectiveness of the BayesDiff procedure relative to existing techniques for differential DNA methylation. Returning to the motivating dataset, we detect the genomic signature for four subtypes of colorectal cancer.