Methodology for inference of intercellular gene interactions

Abstract

To govern organ size, shape, and function, cell-secreted diffusible molecules called morphogens spatially pattern cell differentiation, gene expression, and proliferation. Local morphogen concentration governs cell differentiation through gene regulatory networks (GRN). Previous inference methodologies tackle intercellular GRN inference between cells of one type. This is insufficient, as many developmental systems consist of cells of different types interacting with each other. Inference methodologies of GRNs between different cell types assume knowledge of diffusible morphogen identity and concentration. This makes their applicability limited in real biological systems. Here, we develop a computational methodology to infer the intercellular GRN derived from experiments that use fluorescence from reporter proteins for gene expression measurements. For validation, we demonstrate the methodology in silico using three case studies based on developmental and synthetic biology. The results show that, barring practical identifiability limitations, the methodology successfully infers the intercellular GRNs.