Gene regulatory networks are complex networks composed of nodes representing genes, transcription factors, microRNAs and other components or modules and their mutual interactions represented by edges. These networks can reveal and depict the fundamental gene regulatory mechanisms in the cells. In this paper we compare the obtained results of gene regulatory networks inference from gene expression microarray data. We have used Dynamic Bayesian networks, Boolean networks and graphical Gaussian models as models for network inference. We applied three different size gene expression datasets simulated by using a simple autoregressive process. After network inference, we compared the values of the area under ROC curve (AUC) as a validation measure. Some directions for further improved approach for GRNs reconstruction which will include prior knowledge are proposed at the end of this paper.
gene regulatory networks, Boolean networks, graphical Gaussian models, area under ROC curve, Bayesian networks, bioinformatics