A clustering-based approach for inferring recurrent neural networks as gene regulatory networks

Constructing genetic regulatory networks is one of the most important issues in system biology research. Yet, building regulatory models manually is a tedious task, especially when the number of genes involved increases with the complexity of regulation. To automate the procedure of network construction, in this work we establish a clustering-based approach to infer recurrent neural networks as regulatory systems. Our approach also deals with the scalability problem by developing a clustering method with several data analysis techniques. To verify the presented approach, experiments have been conducted and the results show that it can be used to infer gene regulatory networks successfully.