Structure of Cortical Circuits
We are developing games for connectomics -- explore the brain in your browser: BRAINFLIGHT. Coming soon...
In our lab, we want to understand the structure of neocortical circuits.
Our goal is to measure the connectivity between thousands of nerve cells in the cerebral cortex, and to understand the computations that this network can perform.
More specifically, we are studying a piece of mouse somatosensory cortex, which analyzes the contact between the main whiskers on the snout of the animal and real-world objects (trees, apples, other mice).
Please see below for a more detailed introduction.
The dense reconstruction of neuronal circuits has become possible with the invention of the serial blockface scanning electron microscope (SBEM, see publication in PLoS Biology 2004).
This new microscope combines a scanning electron microscope to efficiently image the top of a block of nerve tissue at very high resolution with a device to shave off an extremely thin layer (~25 nm thickness) from the top of that tissue block. All of this is completely automated. These experiments run for weeks and months, and are therefore very challenging.
Once the images are acquired, the data has to be reconstructed. That is, the nerve cells with all their elaborate processes (the wires of the brain) have to be extracted, and synapses have to be detected.
Based on tools developed by Helmstaedter, Briggman and Denk (Nature Neuroscience 2011), we are able to crowd source the reconstruction task, and thus reconstruct circuits of considerable size.
For high-throughput nerve cell reconstruction, we use a combination of humans (currently more than 100 undergraduate students) and machines.
The main part of the human (and most mammalian) brains is the outer shell directly beneath the skull, called the cerebral cortex (or neocortex). This is where we see, hear, and think.
The neocortex is packed with nerve cells. Even a small piece the size of, say, a grain of sand, contains several thousands of neurons. What is most stunning, though, is not the sheer number of nerve cells (the liver also contains a lot of cells), but the extreme level of connectivity between them.
Nerve cells are like humans: they talk directly to at least hundreds, if not thousands of their kind. The network of neurons is very complex, and it is very likely that it is this extreme complexity that makes the brain so powerful.
Yet, we don't know what the connectivity between large clusters of nerve cells in the cortex looks like.
Therefore, our goal is to measure the connectivity between thousands of neurons in the cerebral cortex, and to understand the computations that this neuronal network can perform.