Oxford University Press's
Academic Insights for the Thinking World

The future of systems neuroscience

How does the brain work? It’s a question on a lot of people’s minds these days, especially with the launch of massive new research efforts like the American BRAIN Initiative and the European Human Brain Project. It’s also a systems question because after all, the brain is a key part of the nervous system, like the skull is a key part of the skeletal system or the heart is a key part of the circulatory system. The basic approach to understanding how any system works has been clear since Greek and Roman times two thousand years ago: understand what the system does, make a parts list, describe how each part works, and then determine how the parts interact to carry out the various functions of the system.

Science is based on observing nature and testing resulting hypotheses to understand functional mechanisms. And major progress comes from the most general hypotheses—theoretical frameworks at the systems level: paradigms. Famous examples include Copernicus organizing the sun and planets with the earth rather than the sun at the center, Mendeleev arranging the basic chemical elements of all matter into a periodic table, and Watson and Crick’s model of the molecular basis of heredity—in terms of how the four nucleotide building blocks of DNA are arranged spatially in a double helix.

Systems neuroscience does not have a comparable theoretical framework, leaving it in a pre-Watson and Crick, or maybe even better, a pre-Darwin state of affairs. The solution is simple, obvious, and attainable—but essentially ignored in current “big science” approaches to neuroscience. Watson and Crick’s model of DNA led 50 years later to the sequencing of the human genome. At first, this project was widely criticized as frivolous, but it proved to be seminal in many ways, not the least of which is establishing the scope of the problem—the basic overall organization of the chromosomes—and allowing the relatively fast and cheap assaying of genome-wide expression patterns on a tiny chip. Getting the structural sequence was only the first step, but it was a necessary step, allowing all of functional, mechanistic understanding to follow logically, in a classic hypothesis-driven way.

Stained human neocortical pyramidal cell. Image credit: Bob Jacobs, Laboratory of Quantitative Neuromorphology Department of Psychology Colorado (CC BY-SA 3.0).
Stained human neocortical pyramidal cell. Photo by Bob Jacobs, Laboratory of Quantitative Neuromorphology Department of Psychology Colorado. CC BY-SA 3.0 via Wikipedia.

The analogous solution for neuroscience is figuring out the basic wiring diagram of the nervous system, and this has to start with the connectome, essentially a table of connections between the parts. From this connectome a blueprint of the nervous system can be developed, like the architectural drawings for an office building, the plans for an airliner, or the schematics for a motherboard. The basic circuit diagram is like a skeleton, a basic framework for understanding the function of the nervous system. It is hard to imagine building and fixing a modern skyscraper, airplane, or computer without detailed and accurate schematics—and the same applies to understanding mechanisms underlying brain function and fixing problems scientifically.

Everybody knows that the brain is the most complex object on earth, so a viable strategy for solving the wiring diagram is essential. The approach here is also obvious—start with the simplest level of analysis, and progress to deeper and deeper levels. The simplest level is the wiring diagram between basic parts, and there are about 500 of them in mammals. This is analogous to displaying the airline routes between major cities around the world. The next level is the wiring diagram at the level of neuron types that make up each part—there are probably 2,500 to 5,000 neuron types in mammals. And the next level after that is the wiring diagram between all of the individual nerve cells that make up each of the neuron types—hundreds of millions to billions in mammals. The simplest, most general level can be solved now with current technology in rodents. Why not do it, and develop more efficient technology at the same time—just like the history of the genome project. Developing effective ways to interact between animal connectomes based on histology at cellular resolution and human connectomes based on MRI—and correlating both with genomic information—is the wave of the future. Great progress in diagnosing, treating, and understanding the etiology of nervous system diseases can be expected by correlating the results of genome-wide association studies with connectome-wide association studies.

Headline image credit: Diagram of brain synapses. Image by Allan Ajifo, aboutmodafinil.com. CC BY 2.0 via Wikimedia Commons.

Recent Comments

There are currently no comments.