What is a connectome, anyway?
We asked on Quora and David Zhou, Masters Student at Carnegie Mellon University, answered.
A connectome is a wiring diagram of the brain.
Two brain cells, or neurons, are “wired” together when parts of their cell bodies communicate with one another through the extracellular secretion and reception of chemicals known as neurotransmitters.
A wiring diagram at the single cell level depicts every connection between two of the brain’s approximately 100 billion neurons. Each neuron makes approximately [at least] a thousand connections to other neurons.
A wiring diagram at the tissue or structural level depicts the size and strength of nerve bundles that travel between brain regions.
Currently, the best way to see these single cell connections is to fix whole brain tissue, slice off layers just a few microns thick, image each slice with an electron microscope, and trace the path of each neuron. The best way to see larger, structural connections is using diffusion magnetic resonance imaging, which applies a magnetic field in order to image the path of water movement inside nerve fibers and tracts, and in doing so, the path of the fibers themselves.
These connections do not exist randomly. They are built, enhanced, pruned, and weakened during fetal development and over the course of a person’s lifetime. They are modulated through human development and experience to store memories, modulate responses, calibrate movement, and compute thought.
A brain can be represented by its connectome, but it is also much more.
If the brain were a subway system, its neurons stations, and its intracellular processes tracks between stations, its connectome would be a full map of the system.
A connectome is a structural diagram. It can depict anatomy, but cannot immediately explain function. Just as a subway map can give you a path between two places, it cannot tell when the trains run, who is likely to travel over them, or what purpose those trains serve the greater city.
Not only is a connectome an abstraction of the biological microstructure of a brain, it is also a data structure for representing the storage of information in a brain.
To researchers, a connectome can be represented by a graph, which is a set of nodes with connections, or edges, in between them. As a graph, the connectome can be represented by data structures commonly used by computer scientists, such as adjacency matrices or adjacency lists. Each of these data structures assigns a numerical value to the presence or strength of each connection between two neurons – a from neuron and a to neuron. The totality of encoded information stored in the brain can, in theory, be recorded as a very large assembly of numbers.
Still, the biological mechanisms for information encoding, processing, and retrieval are unclear. This makes it difficult to say the shape that information in the brain takes. Is information in the brain discretely stored or distributed? Does it have to be passed along or handled by preexisting signals travelling across these neurons? If a connectome is a data structure, it is not one that we can currently read or access.
Connectomics are (part of) the future to a full understanding of brain function.
No two person’s connectomes are identical, because no two person’s past experiences, genetic makeup, and brain development are the same. Nevertheless, there are large-scale trends that human connectomes have in common. Most human brains have similar structures in similar places with a stereotypical configuration of cells (for example, in layers.)
Something we want to know about the connectome is not only how they differ between humans, but also how they don’t. Something else we want to know about the connectome is which differences capture variation in thought, personality, and memory – and which capture variation in mental health and disease.
Some neuroscientists are very hopeful about the insights that can be gleaned from such a detailed picture of the brain. They hope that we may perform single cell computer simulations of brain circuits, regions, or even the brain whole. They hold in the connectome hope for the physical substrate of brain phenomena that biologists have not yet been able to touch: learning, emotion, cognition, memory, and awareness.
Other neuroscientists are more hesitant. They note that we have enough unsifted data in neuroscience research. They argue that we can make a lot of progress with what we already have without pouring our hopes into what might be an intractable project. They argue that a lot of emergent or complex phenomena will still be out of reach when we have mapped the connectome.
Reservations notwithstanding, there is very little in neuroscience that will not be impacted by a full mapping of neuronal connections.