We've been talking a lot this week about networks, but mostly of the tube-ed, digital internet variety. Another exciting network is of the squishy, pink, biological variety – the brain. Some neuroscientists have recently taken it upon themselves to coin a new label for a complete brain network, the "connectome."
WTF is a "connectome?"
Small teams of neuroscientists at universities across the world have, of late, gravitated to this young field of neuroscience, called "connectomics." Researchers at both Harvard and MIT, led by, respectively, neuroscientists Jeff Lichtman and Sebastian Seung, are spearheading several ambitious new connectomics projects, working toward the complete mapping of the human brain and all of its synaptic connections. They coined the term "connectome" to describe this hypothetical complete map, in reference to the genome – the map of our genes. A connectome is a complete, detailed, and, at the moment, theoretical map of all of your billions of neurons and their billions of connections.
The scope of the project is almost unfathomable, and the sexiness of connectomics has garnered a lot of public attention. A complete connectome of a human brain would be an important landmarks in the entire field of neuroscience.
Inevitably, a project of this reach faces skepticism, just like the other large-scale brain mapping project, Henry Markram's controversial, partially IBM-funded, Blue Brain. While the connectome endeavor has received more general support from the neuroscience community than Blue Brain, criticism of both projects has a similar theme.
The fact is, we barely understand the basic biology of the most simple neural circuits. Activity at every synapse is so incredibly varied, changes so incredibly fast, and often takes decades for neuroscience to even begin to grasp. For instance, I spent two years researching the neural circuit in the mouse cerebellum responsible for the timing of eye blinks when air hits the cornea (why can't there be a TED talk about that!?). Naturally, we celebrated even the smallest of clues, and were most definitely not at the point of understanding the myriad details of each step of the basic physiological process. So, the criticism goes, a mapping of the brain and its connections, and even the activity and short-term change in each connection (which Seung and his colleagues hope to eventually map as well), is impossible to achieve at this moment without a serious simplification of neural dynamics. To be accurate, the project would have to constantly filter and encompass new (and often conflicting) data on the molecular doings of specific neural circuits – data which are published every day.
However ambitious they seem, the connectomics folks don't plan to finish the project in their lifetimes, as Seung says in the TED clip above. Supporters of the project make a strong case that we need projects like this to get scientists to think about the "big picture," and start developing the technology to eventually see these "big" projects through. Jeff Lichtman, of Harvard, has already started doing so, with his crafty "Brainbow" staining technique.
Lichtman's technique uses a combination of electron microscopy, brain slicing, and fluorescent dying to see neuron connections in detail. Fluorescent microscopy works by labeling cells with specific markers that cause them to glow certain colors when bathed in a special wash of chemical agents (fluorophores). These "markers" are usually genetic markers, and by tinkering with the genome of a host animal, the markers – and thus the colors produced by cells under the microscope – can be altered. Lichtman has honed the technique and has over one hundred colors in his palette. Here's a picture from his lab, showing a portion of a mouse brain (the dentate gyrus) and its labeled connections:
Other techniques, in computer modeling and data storage, will bolster the arsenal of connectomics. But a complete connectome, when/if it ever exists, would not just be an impressive example of cutting-edge biotech, says Seung, it could also represent the personalilty, thoughts, and feelings of whomever's connectome is cracked. Given enough detail, the theory goes, a connectome could give a snapshot of a self:
Since the connectome defines the pathways along which neural activity can flow, we might regard it as the streambed of consciousness. The metaphor is a powerful one. Over a long period of time, in the same way that the water of the stream slowly shapes the bed, neural activity changes the connectome. The two notions of the self — as both the fast-moving, ever-changing stream, and the more stable but slowly transforming streambed — are thus inextricably linked.
Connectomics is partially science fiction at the moment, but, as Freeman Dyson recently wrote, "All of science is uncertain and subject to revision. The glory of science is to imagine more than we can prove."