|
Abstract:
|
A major research issue in human microbiome studies is the feasibility of designing interventions that modulate the composition of the microbiome to promote health and cure disease. This requires extensive understanding of the microbiome’s modulating factors, such as dietary intake, and the relation between microbial composition and phenotypic outcomes, such as body mass index. Previous efforts have modeled these data separately, with two-step approaches that can produce biased interpretations of the results. We propose a Bayesian joint model that jointly identifies clinical covariates associated with microbial composition data and predicts a phenotypic response using information in the compositional data. We apply our model to understand relations between dietary intake, microbial samples, and BMI. We find numerous associations between microbial taxa and dietary factors that may lead to a microbiome that is more hospitable to the development of chronic diseases such as obesity.
|
