Deep pathway analysis incorporating mutation information and gene expression data

Abstract

We propose a new way of analyzing biological pathways in which the analysis combines both transcriptome data and mutation information and uses the outcome to identify routes of aberrant pathways potentially responsible for the etiology of disease. Each pathway route is encoded as a Bayesian Network which is initialized with a sequence of conditional probabilities which are designed to encode directionality of regulatory relationships encoded in the pathways, i.e. activation and inhibition relationships. First, we demonstrate the effectiveness of our model through simulation in which the model is able to discern patients in Test Group from ones in Control Group. Second, we apply our model to analyze the Breast Cancer data set, available from TCGA, against some pathways available from KEGG. Our experiment with this published data reaffirms the claims reported from the original breast cancer PAM50 subtype study. Our model can further analyze the patients of each subtype based on the identified route of aberration. For example, our analysis shows that complex biological process patterns are presented for HER2+ patients potentially suggesting our method's use for producing refined subtyping. We manage to find commonly perturbed pathway routes for HER2+ patients. We claim such “deep” pathway analysis could be very useful in designing a personalized therapy.