IARPA awards $18.7M contract to Allen Institute to reconstruct neuronal connections
The Intelligence Advanced Research Projects Activity (IARPA) has awarded an $18.7 million contract to the Allen Institute for Brain Science, as part of a larger project with Baylor College of Medicine and Princeton University, to create the largest ever roadmap to understand how the function of networks in the brain's cortex relates to the underlying connections of its individual neurons.
The project is part of the Machine Intelligence from Cortical Networks (MICrONS) program, which seeks to revolutionize machine learning by reverse-engineering the algorithms of the brain.
"This effort will be the first time that we can physically look at more than a thousand connections between neurons in a single cortical network and understand how those connections might allow the network to perform functions, like process visual information or store memories," says R. Clay Reid, Ph.D., Senior Investigator at the Allen Institute for Brain Science, Principal Investigator on the project.
The ultimate goal is to implement the algorithms and learning rules that scientists decipher from the brain to advance the field of artificial intelligence. Artificial neural networks have recently developed capabilities to do speech recognition, recognize faces and help analyze big data for biomedical research. "However, in many ways, these artificial neural networks are still primitive compared to biological networks of neurons and do not learn the way real brains do," says Andreas Tolias, Ph.D., Associate Professor in the Department of Neuroscience at Baylor College of Medicine. "Our goal is to fill this gap and apply the algorithms of the brain to engineer novel artificial network architectures."
Experiments for the project will begin at Baylor College of Medicine, where researchers will perform studies to identify neurons that respond to particular patterns of visual stimulation, such as lines in different orientations on a screen.
The experiments will continue at the Allen Institute for Brain Science, where sections of the same brains will be stained, sliced and imaged in an array of electron microscopes, which capture the intricate details of neurons and their connections.
"The data we're collecting is massive, with the work having virtually no margin for error," says Christof Koch, Ph.D., President and Chief Scientific Officer of the Allen Institute for Brain Science. "If we lose two slices of brain in a row, each of which is vastly thinner than a human hair, we cannot hope to reconstruct the segment faithfully in three dimensions and have to start over. This is where the Allen Institute's history of processing and scaling data effectively at an industrial scale is so crucial."
Finally, the 3D image data will be sent to Princeton University, under Sebastian Seung, Ph.D., Professor of Computer Science and the Princeton Neuroscience Institute, where it will be painstakingly reconstructed in three dimensions by human annotators aided by powerful machine vision and machine learning algorithms, and each individual neuron with all its myriad processes will be traced and analyzed.
"We are delighted to be collaborating on this exciting project with our superb scientific colleagues at Baylor and the Allen Institute," says Seung.
The end goal of the project is to create a reconstruction of a cubic millimeter of brain tissue, the size of a grain of sand, yet containing the largest section of brain ever to be studied in this way to date.
"This work is the next big step in understanding how networks in the brain work," say Reid. "For the first time we will be able to come to grips with cortical networks in all their complexity, rather than a little bit at a time. No other technique can hope to grapple with their richness."
Provided by Allen Institute for Brain Science
