A Bayesian Network Approach to Disease Subtype Discovery

Food co-consumption network as a new approach to dietary pattern in non-alcoholic fatty liver disease

Dietary patterns strongly correlate with non-alcoholic fatty liver disease (NAFLD), which is a leading cause of chronic liver disease in developed societies. In this study, we introduce a new definition, the co-consumption network (CCN), which depicts the common consumption patterns of food groups through network analysis. We then examine the relationship between dietary patterns and NAFLD by analyzing this network. We selected 1500 individuals living in Tehran, Iran, cross-sectionally. They completed a food frequency questionnaire and underwent scanning via the FibroScan for liver stiffness, using the CAP score. The food items were categorized into 40 food groups. We reconstructed the CCN using the Spearman correlation-based connection. We then created healthy and unhealthy clusters using the label propagation algorithm. Participants were assigned to two clusters using the hypergeometric distribution. Finally, we classified participants into two healthy NAFLD networks, and reconstructed the gender and disease differential CCNs. We found that the sweet food group was the hub of the proposed CCN, with the largest cliques of size 5 associated with the unhealthy cluster. The unhealthy module members had a significantly higher CAP score (253.7 ± 47.8) compared to the healthy module members (218.0 ± 46.4) (P < 0.001). The disease differential CCN showed that in the case of NAFLD, processed meat had been co-consumed with mayonnaise and soft drinks, in contrast to the healthy participants, who had co-consumed fruits with green leafy and yellow vegetables. The CCN is a powerful method for presenting food groups, their consumption quantity, and their interactions efficiently. Moreover, it facilitates the examination of the relationship between dietary patterns and NAFLD.

From heterogeneous healthcare data to disease-specific biomarker networks: A hierarchical Bayesian network approach

In this work, we introduce an entirely data-driven and automated approach to reveal disease-associated biomarker and risk factor networks from heterogeneous and high-dimensional healthcare data. Our workflow is based on Bayesian networks, which are a popular tool for analyzing the interplay of biomarkers. Usually, data require extensive manual preprocessing and dimension reduction to allow for effective learning of Bayesian networks. For heterogeneous data, this preprocessing is hard to automatize and typically requires domain-specific prior knowledge. We here combine Bayesian network learning with hierarchical variable clustering in order to detect groups of similar features and learn interactions between them entirely automated. We present an optimization algorithm for the adaptive refinement of such group Bayesian networks to account for a specific target variable, like a disease. The combination of Bayesian networks, clustering, and refinement yields low-dimensional but disease-specific interaction networks. These networks provide easily interpretable, yet accurate models of biomarker interdependencies. We test our method extensively on simulated data, as well as on data from the Study of Health in Pomerania (SHIP-TREND), and demonstrate its effectiveness using non-alcoholic fatty liver disease and hypertension as examples. We show that the group network models outperform available biomarker scores, while at the same time, they provide an easily interpretable interaction network.

High-dimensional and heterogeneous healthcare data, such as electronic health records or epidemiological study data, contain much information on yet unknown risk factors that are associated with disease development. The identification of these risk factors may help to improve prevention, diagnosis, and therapy. Bayesian networks are powerful statistical models that can decipher these complex relationships. However, high dimensionality and heterogeneity of data, together with missing values and high feature correlation, make it difficult to automatically learn a good model from data. To facilitate the use of network models, we present a novel, fully automated workflow that combines network learning with hierarchical clustering. The algorithm reveals groups of strongly related features and models the interactions among those groups. It results in simpler network models that are easier to analyze. We introduce a method of adaptive refinement of such models to ensure that disease-relevant parts of the network are modeled in great detail. Our approach makes it easy to learn compact, accurate, and easily interpretable biomarker interaction networks. We test our method extensively on simulated data as well as data from the Study of Health in Pomerania (SHIP-Trend) by learning models of hypertension and non-alcoholic fatty liver disease.