BayeSuites: An open web framework for massive Bayesian networks focused on neuroscience

Abstract

<jats:title>Abstract</jats:title><jats:p>BayeSuites is the first web framework for learning, visualizing, and interpreting Bayesian networks (BNs) that can scale to tens of thousands of nodes while providing fast and friendly user experience. All the necessary features that enable this are reviewed in this paper; these features include scalability, extensibility, interoperability, ease of use, and interpretability. Scalability is the key factor in learning and processing massive networks within reasonable time; for a maintainable software open to new functionalities, extensibility and interoperability are necessary. Ease of use and interpretability are fundamental aspects of model interpretation, fairly similar to the case of the recent explainable artificial intelligence trend. We present the capabilities of our proposed framework by highlighting a real example of a BN learned from genomic data obtained from Allen Institute for Brain Science. The extensibility properties of the software are also demonstrated with the help of our BN-based probabilistic clustering implementation, together with another genomic-data example.</jats:p>

SAMBA: Structure-Learning of Aquaculture Microbiomes using a Bayesian Approach

“…However, our understanding about the dynamics of interrelations between fish microbiomes and their hosts is still in its infancy due to the multiple biotic and abiotic factors involved in this kind of dynamics (Faust, 2021;Liu et al, 2021). Certainly, there is now a well-known recognized interest for system biology tools like Bayesian networks (BN) and structure learning systems (Scutari, 2009;Michiels et al, 2021) that can be convincingly used to realistically modelize complex biological systems. In the field of Microbiology, BNs have proven to be a powerful tool for inferring directional relationships within microbial communities (Sazal et al, 2020;Sazal et al, 2021), and for analyzing functional networks in metagenomics data (Hobbs et al, 2016).…”

“…Current state-of-the-art in BNs includes several tools with interactive visualizations, such as ShinyBN (Chen et al, 2019) or BayesiaLab (Conrady and Jouffe, 2015), which are only able to work with discrete variables and small datasets. The recent BayeSuites tool (Michiels et al, 2021) overcomes these difficulties, being capable of managing continuous variables and large datasets focused on neurosciences, albeit it is not able to perform inference on discrete variables to establish their conditional probability distributions. In aquaculture research, BNs have been used to better integrate knowledge on sustainable ecosystem-based management (Yuniarti et al, 2021), but the structure of interrelations and dependencies among the multiple distinct biotic and abiotic factors involved in the dynamics of a given aquaculture system remains mostly unknown.…”

Dynamic evaluation of induced epilepsy in rats: a bayesian network perspective

“…However, the objective of this thesis is to understand these relationships and structure from data, which has been reached using a Dynamic Bayesian Network. A Bayesian Network (BN) is a compact representation of statistical dependencies among variables (MICHIELS; LARRAÑAGA; BIELZA, 2021;LARRANAGA, 2014;KOLLER;FRIEDMAN, 2009;NEAPOLITAN et al, 2004). BNs are probabilistic models defined by a Directed Acyclic Graph (DAG) and conditional probabilities tables (CPT) representing the probabilistic dependence over signals.…”