This award to Rochester Institute of Technology is to develop novel computational methods to better reconstruct gene regulatory networks using metabolic simulation. Accurate prediction of the behavior of gene regulatory networks is necessary to advance areas such as biotechnology, drug development, and whole cell modeling, and the integration of prior biological knowledge about regulatory structure and/or behavior is critical for successful predictions. However, using prior metabolic knowledge is challenging, as it is indirect and can only be evaluated using metabolic simulation. In this project, the team is developing computational methods that will allow the metabolic prior to be integrated into the regulatory network reconstruction.
The advances made in this study will improve the accuracy of gene regulatory network reconstruction, and will help to determine to what degree integration of metabolic models can improve regulatory network reconstruction. The interdisciplinary research team combines expertise in large-scale computationally intensive statistical modeling and molecular biology, and is well equipped to assure the validity, tractability and statistical soundness of the work. Several students will be trained in the activities, representatives from underrepresented groups will be included, and the outcomes will include software tools that are freely available to the research community.
