From social networks to neurosciences, graphs have rapidly become ubiqui- tous by offering a versatile modeling framework in which data points are repre- sented as nodes, and various aspects of the underlying organization of the data are captured through edges. Brain Connectomics ––– a developing field in cognitive neuroscience ––– is a case in point, as it strives to understand cognitive processes and psychiatric diseases through the analysis of interactions between brain regions. However, in the high-dimensional, low-sample, and noisy regimes that typically characterize fMRI data, the recovery of such interactions remains an ongoing challenge: how can we discover robust patterns of co-activity between brain regions that could then be associated to cognitive processes or psychiatric disorders? How can we quantify the uncertainty associated to these discoveries? In this talk, we investigate a constrained Bayesian Independent Component Analyis (ICA) approach which simultaneously allows (a) the flexible integra- tion of multiple sources of information (fMRI, DWI, anatomical, etc.), (b) an automatic and parameter-free selection of the appropriate sparsity level and number of connected submodules and (c) the provision of estimates on the un- certainty of the recovered interactions. Our experiments, both on synthetic and real-life data, validate the flexibility of our method and highlight the benefits of integrating anatomical information for connectome inference.