Inferring Gene Regulatory Networks from Multiple Data Sources Via a Dynamic Bayesian Network with Structural EM

摘要: Using our dynamic Bayesian network with structural Expectation Maximization (SEM-DBN), we develop a new framework to model gene regulatory from both expression data and transcriptional factor binding site data. Only based on mRNA data, it is not enough accurately estimate network. It difficult for us only the In this paper, use transcription location in order introduce prior knowledge SEM-DBN model. Gene are also exploited specifically likelihood. Meanwhile, incorporate into every learning step by SEM rather than very beginning, which can compensate attenuation of effect The effectiveness proposed method demonstrated through analysis Saccharomyces cerevisiae cell cycle combination heterogeneous multiple sources ensures that results more accurate those recovered alone.