copula gaussian graphical modelling of biological networks and bayesian inference of model parameters

A proper understanding of complex biological networks facilitates a better perception of those diseases that plague systems and efficient production of drug targets, which is one of the major research questions under the personalized medicine. however, the description of these complexities is challenging due to the associated continuous, high-dimensional, correlated and very sparse data. the copula gaussian graphical model (cggm), which is based on the representation of the multivariate normal distribution via marginal and copula terms, is one of the successful modeling approaches to presenting such types of problematic datasets. this study shows its novelty by using cggm in modeling the steady-state activation of biological networks and making inference of the model parameters under the bayesian setting. in this regard, the reversible jump markov chain monte carlo (rjmcmc) algorithm is suggested in order to estimate the plausible interactions (conditional dependence) between the systems' elements, which are proteins or genes. furthermore, the open-source r codes of rjmcmc are generated for cggm in different dimensional networks. in this regard, real datasets are applied, and the accuracy of estimates via f-measure is evaluated. from the results, it is observed that cggm with rjmcmc is successful in presenting real and complex systems with higher accuracy.