Bayesian Gaussian Copula Graphical Modeling for Dupuytren Disease

Abstract

Dupuytren disease is a fibroproliferative disorder with unknown etiology that often progresses and eventually can cause permanent contractures of the affected fingers. In this paper, we provide a computationally efficient Bayesian framework to discover potential risk factors and investigate which fingers are jointly affected. Our Bayesian approach is based on Gaussian copula graphical models, which are one potential way to discover the underlying conditional independence structure of variables in multivariate mixed data. In particular, we combine the semiparametric Gaussian copula with extended rank likelihood which is appropriate to analyse multivariate mixed data with arbitrary marginal distributions. For the graph structure learning, we construct a computationally efficient search algorithm which is a trans-dimensional MCMC algorithm based on a birth-death process. In addition, to make our statistical method easily accessible to other researchers, we have implemented our method in C++ and interfaced with R software as an R package BDgraph which is available online.