On Learning Gene Regulatory Networks Under the Boolean Network Model
Machine Learning
Dynamic bayesian networks: representation, inference and learning
Dynamic bayesian networks: representation, inference and learning
Inferring Connectivity of Genetic Regulatory Networks Using Information-Theoretic Criteria
IEEE/ACM Transactions on Computational Biology and Bioinformatics (TCBB)
Hi-index | 0.00 |
Inferring gene regulatory networks (GRNs) is a challenging inverse problem. Most existing approaches have understandably low accuracy because of the intrinsic complexity of a biology system and a limited amount of available data. Dynamic Bayesian Network (DBN) is one of the widely used approaches to identify the signals and interactions within gene regulatory pathways of cells. It is well-suited for characterizing time series gene expression data. However, the impacts of network topology, properties of the time series gene expression data, and the number of time points on the inference accuracy of DBN are still unknown or have not been fully investigated. In this paper, the performance of DBN is evaluated using both in-silico yeast data and three growth phases of Yeast Saccharomyces cerevisiae cell cycle data with different time points in terms of precision and recall. The inferred GRNs were compared with the actual GRNs obtained from SGD (The Saccharomyces Genome Database) in terms of precision and recall. This work may provide insight and guideline for the development and improvement of GRN inference methods.