How mapping a brain could change the world.
No-one today would question the global worth of the Human Genome Project, a complex and expensive plan to map all the DNA in a human being. Its success has revolutionised modern medicine.
Yet when the project began in 1990, critics questioned the UK’s decision to spend such a large amount of money on research that, ‘other countries will do anyway’. Even the father of DNA, James Watson said ‘why do the human when we haven’t done the bacteria yet’. Watson soon changed his mind when he realised the scientific potential it would unleash.
By ignoring sceptics and helping to fund the project, the government of the time invested in the UK’s future status as a science superpower. Our scientists at the MRC Lab of Molecular Biology (LMB) and the Sanger Institute spearheaded the research, and Britain is still reaping huge economic, health and reputational rewards today.
Now, the same prestigious Cambridge lab that helped invent modern genomics – the LMB – wants to focus on making the next map with the potential to transform global health, a map of the whole brain, called a connectome.
the Flywire Consortium, which includes researchers from the University of Cambridge and the
MRC Laboratory of Molecular Biology
The Connectome
Just as a genome is a complete map of one organism’s DNA, so a connectome is a complete, wiring map of one brain.
To give you an idea of just how complex and ambitious this project is, the map of a mouse’s connectome would be 100,000 bigger than the map of its genome. The raw 3D image data alone would fill a half a million laptops. And that’s just a mouse, imagine the size of a human connectome.
Yet this is critical research.
Our lack of knowledge on how the brain works is a major barrier to understanding the mechanics of dementia and mental illness. It is like a computer scientist trying to solve a software problem without being able to see any of the code. Some mental health issues could be down to neurons making wrong connections or, conversely, not making the right connections. But without a connectome, scientists have a hard time determining who might be talking to whom.
The LMB already has the pedigree to lead the field in this work. Its scientists invented connectomics in the 1970s, successfully mapping the brain connections of a nematode worm. Their more recent research on fly brains has proved the principle, giving clues to the neural substrate behind making memories, aggression, taking-decisions and navigating spaces.
It is now time to move onto mapping more complex and important brains.
The important thing to keep in mind when it comes to connectomes is that they are, in the first place, anatomical descriptions of neurons and their physical connections. A connectome alone may not tell you much about how a given circuit functions but it will help you to constrain the problem and design experiments to figure that out.
How investing in the Human Genome Project is paying the UK back many times over
As a direct result of its genomic leadership, Britain has attracted substantial investment from global pharmaceutical companies such as AstraZeneca, GlaxoSmithKline and Johnson & Johnson. It is no coincidence that AstraZeneca recently built its £1bn R&D headquarters next to the MRC Lab of Molecular Biology and down the road from the Sanger Institute, both pivotal players in the Human Genome Project and leaders in current genetic research.
Britain also has the world’s most detailed genomics database, the UK Biobank, which is a powerful tool for global research, aiding in the understanding, prevention, and treatment of diseases. The ability to link genetic data with health outcomes has accelerated drug discovery and development, with genetic insights doubling the success rate of new treatments.
Revolutionising AI
Connectomics also has the potential to revolutionise AI. The most recent step-change in that technology – Deep Mind – came from work to model AI on the human brain. The models currently used in AI are known to be orders of magnitudes less efficient than brains and the network models are only very loosely modelled on what could be happening in our brains. Being able to model that much more accurately opens up the prospect of being able to simulate, for example, what really happens in mental disorders.
