Autism spectrum disorder (ASD) is a complex neurodevelopmental condition that is the result of interplay amongst hundreds of genes. Even though researchers have implicated dozens of genes in ASD to date, there is still much more to be understood. A few network-based ASD gene discovery algorithms have been developed with the following goals: (1) speeding up the gene discovery process by using the guilt-by-association principle, and (2) understanding affected functionalities by analyzing genes and links in predicted clusters. Fundamentally, these algorithms rely on the assumption that ASD risk genes are part of a functional gene network. However, the definition of a gene network in these analyses is a single, flat and static network. This approach disregards the temporal dimension in the development and differentiation of neurons and brain tissue. The assumption of ASD genes being part of a functional network is reasonable. However, the functional clustering of genes is bound to evolve over time and more than likely to have a cascading effect on future associations.
A. Ercument Cicek and his team plan to design and develop algorithms that investigate dynamic gene interaction networks across time. They will test the hypotheses that ASD risk genes are clustered throughout the course of neurodevelopment, rather than just at a single snapshot in time, and that functional circuits are sequentially affected. The goal of the project is to discover genes and functionalities that are (1) overlooked, as they are active at less interesting times or in less interesting brain regions and (2) that are affected throughout neurodevelopment.