Discovers genome-phenome relationship using Bayesian Networks (BNs). MBS employs the extended greedy search and learn directed acyclic graph (DAG) models that contain two or more predictors in the epistatic interaction. It can be used to learn from data the interactive relationship among a subset of predictors that together can have a causal effect on a clinical feature. This tool has been tested on genome-wide association study (GWAS) data and successfully discovered the epistatic interaction of single nucleotide polymorphisms (SNPs) that have causal effect on Late Onset Alzheimer disease (LOAD).
Department of Biomedical Informatics, University of Pittsburgh, Pittsburgh, PA, USA
MBS funding source(s)
Supported by grant R01LM011663 awarded by the National Library of Medicine of the National Institutes of Health, and by a CURE grant awarded by the Pennsylvania Department of Health (PA DOH 4100070287).