Biological Network Inference With GRASP: A Bayesian Network Structure Learning Method Using Adaptive Sequential Monte Carlo

Abstract

Bayesian networks (BNs) provide a probabilistic, graphical framework for modeling high-dimensional joint distributions with complex correlation structures. BNs have wide applications in many disciplines, including biology, social science, finance and biomedical science. Despite extensive studies in the past, network structure learning from data is still a challenging open question in BN research. In this study, we present a sequential Monte Carlo (SMC)-based three-stage approach, GRowth-based Approach with Staged Pruning (GRASP). A double filtering strategy was first used for discovering the overall skeleton of the target BN. To search for the optimal network structures we designed an adaptive SMC (adSMC) algorithm to increase the quality and diversity of sampled networks which were further improved by a third stage to reclaim edges missed in the skeleton discovery step. GRASP gave very satisfactory results when tested on benchmark networks. Finally, BN structure learning using multiple types of genomics data illustrates GRASP’s potential in discovering novel biological relationships in integrative genomic studies.

Artificial Intelligence for Quantitative Modeling in Drug Discovery and Development: An Innovation and Quality Consortium Perspective on Use Cases and Best Practices

“…reflective of the underlying biological mechanisms in a structured manner (e.g., drug-disease and gene-disease relationships, and protein-protein interactions). Yu et al 92 proposed a Bayesian network structure learning method called GRASP, which uses an adaptive sequential Monte Carlo approach to infer the causal relationships between genes. GRASP was able to identify causal relationships between genes that were not previously known, demonstrating its potential in constructing the biomedical KG.…”

“…NLP is important for relation extraction in constructing KGs reflective of the underlying biological mechanisms in a structured manner (e.g., drug‐disease and gene‐disease relationships, and protein–protein interactions). Yu et al 92 . proposed a Bayesian network structure learning method called GRASP, which uses an adaptive sequential Monte Carlo approach to infer the causal relationships between genes.…”