As an embryo develops into an adult, each cell follows a genetic 'script' comprising an intricate network of interactions between regulatory genes. Extensive research has been done to identify these genes and their interactions in order to build a more detailed understanding of how development progresses. Sea urchin embryos are favored models for such research, owing to their manageable size and their transparency, which enables direct visualization of developmental stages.

Biologists at the California Institute of Technology (Pasadena) have focused their research on characterizing the gene interactions that occur during the first 30 hours of development of the endomesoderm in the sea urchin. They have identified ∼50 regulatory genes in this network. Now, for the first time, they have built a computational model of this network, creating a powerful tool for exploring gene regulatory networks (GRNs) in a novel manner.

To construct the model, Isabelle S. Peter, Emmanuel Faure and Eric H. Davidson condensed everything they had learned about the network into a series of logical 'if–then' (or Boolean) statements (e.g., if gene 1 is on, then gene 2 is turned off). By analyzing these sequential statements, the model predicts the location and timing of expression of each gene within the developing embryo (Proc. Natl. Acad. Sci. USA published online 27 August 2012; doi:10.1073/pnas.1207852109). The group compared the predicted expression patterns with experimental observations and found that the model reproduced the data very closely. “It works surprisingly well,” said Peter in a press release.

The team also introduced alterations into the model, such as deactivating or misexpressing certain genes, manipulations that are often done in vivo to assess their effects on development. The model correctly predicted the same results that are seen experimentally in response to such manipulations.

Credit: vilainecrevette at iStockPhoto.com

Peter and her team draw two key conclusions from the success of their computational GRN model. First, their current understanding of the network controlling sea urchin endomesoderm development during the first 30 hours is both thorough and accurate. There were a few inaccuracies in the model's predictions, but these helped to identify pockets of missing or incomplete data. “Identification of these gaps in the GRN model is one of the useful outcomes of the computation,” the authors wrote. Second, their computational approach accurately represents developmental regulation and can be used to test developmental perturbations in silico. Their goal is to similarly model the GRNs underlying each part of a sea urchin embryo during its entire development.