Dr. Dura-Bernal has been awarded a large 1-year NSF grant to develop large-scale cortical models using supercomputers. This grant is part of a program called 'Exploring Clouds for Acceleration of Science' (E-CAS) led by the Internet2 organization and funded by NSF, aimed at demonstrating that commercial cloud computing providers, such as Google Cloud Platform or Amazon Web Services, can be used to advance scientific research. Drs. Dura-Bernal and Lytton's proposal to decipher the brain's neural code through simulations was one of the 6 selected for Phase 1; others included studying data from the Large Hadron Collider (LHC) or the South Pole Neutrino Observatory. In this preliminary stage, each candidate had one year to demonstrate the potential to accelerate science using cloud computing. Dr. Dura-Bernal used supercomputers to explore neural coding mechanisms in motor cortex circuits; they developed new computational methods and ran simulations on 100,000 processors simultaneously (see blog post). A panel of experts then judged their work to decide which of the 6 teams would move on to Phase 2. The SUNY Downstate team, led by Dr. Dura-Bernal, and the MIT LHC team were the winners. Each team received one additional year of funding, including funding for personnel and $450,000 in supercomputing resources. In Phase 2 of this project, the SUNY team aims to build the first biologically-detailed model combining motor cortex, somatosensory cortex and thalamic circuits, in order to investigate the neural coding mechanisms underlying sensorimotor function.
