Bayesian Posterior Integration for Classification of Mass Spectrometry Data

High-throughput technologies currently have the capability to capture information at both global and targeted scales for the transcriptome, proteome, and metabolome, as well as determining functional aspects of these biomolecules. The promise of data integration is that by utilizing these disparate data streams a more accurate predictive model of the phenotype of interest can be developed by identifying the best subset of molecules associated with the outcome. However, in a space of tens of thousands of variables (e.g., genes, proteins), feature selection approaches often yield over-trained models with poor predictive power. Moreover, feature selection algorithms are typically focused on a single source of data and do not evaluate the effect on downstream statistical integration models. The integration of Bayesian statistical outputs have been shown to be an effective approach that optimizes the outcome of interest in the context of the integrated posterior probability. This chapter demonstrates that this approach can improve sensitivity and specificity over simple selection routines based on individual high-throughput datasets generated via mass spectrometry.