Bayesian dynamic network modelling: an application to metabolic associations in cardiovascular diseases

We propose a novel approach to the estimation of multiple Graphical Models to analyse temporal patterns of association among a set of metabolites over different groups of patients. Our motivating application is the Southall And Brent REvisited (SABRE) study, a tri-ethnic cohort study conducted in the UK. We are interested in identifying potential ethnic differences in metabolite levels and associations as well as their evolution over time, with the aim of gaining a better understanding of different risk of cardio-metabolic disorders across ethnicities. Within a Bayesian framework, we employ a nodewise regression approach to infer the structure of the graphs, borrowing information across time as well as across ethnicities. The response variables of interest are metabolite levels measured at two time points and for two ethnic groups, Europeans and South-Asians. We use nodewise regression to estimate the high-dimensional precision matrices of the metabolites, imposing sparsity on the regression coefficients through the dynamic horseshoe prior, thus favouring sparser graphs. We provide the code to fit the proposed model using the software Stan, which performs posterior inference using Hamiltonian Monte Carlo sampling, as well as a detailed description of a block Gibbs sampling scheme.

Automated summary

This paper proposes a novel approach to estimate multiple graphical models for analyzing temporal patterns of association among a set of metabolites across different groups of patients. The study focuses on a tri-ethnic cohort study conducted in the UK, aiming to identify potential ethnic differences in metabolite levels and associations and understand the different risks of cardio-metabolic disorders across ethnicities. Within a Bayesian framework, the authors employ a nodewise regression approach to infer the structure of the graphs, taking into account information across time and ethnicities. The proposed method is implemented using the software Stan, which performs posterior inference using Hamiltonian Monte Carlo sampling.