Inferring Gene Networks from Gene Expression Data Using Dynamic Bayesian Network with Different Scoring Metric Approaches

Abstract

Inferring gene networks can be defined as the process of identifying gene interactions from experimental data through computational analysis. The aim is to infer gene network from gene expression data using dynamic Bayesian network (DBN) with different scoring metric approaches. The previous method, Bayesian network has successfully identified those gene networks but there are some limitations. Hence, DBN is able to infer interactions from a data set consisting time series rather than steady-state data. This research is conducted in order to construct and implement gene network and to analyze the effect by applying a different scoring metric approach for modeling gene network. In order to achieve the goals, a discrete model of DBN is used with different scoring metric approaches which are BDe and MDL. The S. cerevisiae cell cycle pathway is used for this research. To ensure the gene networks are biologically probable, this research employs previous annotation relative to the dataset. By having all of these implementations, this research is able to identify the effect of different scoring metric approaches, identify biologically meaningful gene network within the gene expression datasets and display the results in convenient representations.