Embryonic stem cells (ESC) are capable of differentiating into cells and tissues of all three germ layers, but difficulties in predicting and controlling ESC differentiation limit the development of potential therapeutic applications in regenerative medicine from such cells. Various systematic methods have been used to decode the complex signaling pathways and gene regulatory networks (GRNs) that govern the ESC differentiation, by measuring the expression profiles after perturbing key transcription factors (TFs) [1] and protein–protein interactions (PPI) [2]. Though computational methods have been applied to model the complex dynamics underlying ESC differentiation [3], a systematic method to integrate available large-scale omics data for ESC differentiation and predict the key genes that drive lineage specific differentiation is still lacking.
We developed a novel regression-based network inference method by integrating temporal expression profiles from ESC differentiation, PPI data, and available knowledge regarding to gene functions.
