Scientists Map Entire Mouse Brain Activity

'For a long time, it was common practice for neuroscientists to focus on their favourite brain region,' says Anne Urai, researcher in the Cognitive Psychology unit and head of the Cognitive, Computational and Systems Neuroscience Lab. 'You would build your own device to record activity and develop your own analysis method.' That independent way of working came with a drawback: 'You couldn't be sure whether your findings were replicable, because another scientist would need to repeat your experiment exactly to test that.' As a result, Urai explains, some areas of the brain were never studied at all.

International Brain Laboratory

To change this, the International Brain Laboratory (IBL) was founded in 2017. Urai joined as a postdoc at Cold Spring Harbor Laboratory in New York. The IBL is a collaboration of neuroscientists from twenty universities in Europe and the United States. Inspired by large-scale projects in physics and biology such as CERN and the Human Genome Project, the team set itself the goal of mapping neural activity across the entire brain at cellular resolution.

279 brain areas

'The brain is the most complex structure we know. To understand how it drives behaviour, we needed an international collaboration that could do justice to that complexity,' says Tom Mrsic-Fogel, professor at University College London and IBL member, in a Nature press release. Using standardised tools and analysis methods, twelve different mouse labs succeeded in recording the activity of half a million neurons over seven years, covering 279 brain regions that together account for 95% of the brain's volume.

Whole-brain activity

The project's findings are presented today in two Nature papers. The first investigates what happens in the mouse brain during a decision-making task. In this experiment-set up identically in every lab-the mouse sits in front of a screen where an image appears either on the left or right side. To earn a reward, a sip of lemonade, the mouse must turn a wheel towards the correct side. 'In earlier studies, researchers often focused on the brain's reward system in this process. But in these data we see that decision-making signals are active across the entire brain, not just in a few specific areas,' Urai explains.

A different perspective on the brain

The second paper explores how mice make decisions based on their expectations. If an image repeatedly appeared on the left, for example, the mouse had to estimate whether it would likely appear on the left again. 'We also see that this information is processed throughout the brain-not only in cognitive regions, but also in visual areas and those involved in sensory processing,' says Urai. 'This forces us to look at brain function differently. Neural activity is not confined to a few regions, but is distributed widely across the brain.'

Complex collaboration

Recording brain activity at the level of individual neurons in mice had long been a goal for neuroscientists. Why has it become possible now? Urai: 'First, the probes used to measure brain activity are relatively new.' The Neuropixels probe is a tiny, transparent device, as thin as a human hair. 'These were developed through a challenging partnership between a company and several universities. Now, they can be ordered by labs with clear instructions, making the method much more accessible.'

Another factor is improved expertise in training animals to perform relatively complex tasks. 'Mice are very suitable: you can teach them interesting tasks and simultaneously record their neural activity in great detail. Their brains are also smooth, without folds in the cortex, which makes inserting the probe easier.'

Finally, there is a growing awareness in neuroscience of the value of large-scale collaboration. 'In this team science project, new roles emerged. A project manager coordinated the collaboration, and software developers maintained the databases so that others could easily use them.'

PhD's students

The International Brain Laboratory is making all datasets and software openly available: any researcher can download and use the data for their own questions. Philippa Johnson and Fenying Zang, a postdoc and PhD candidate in Urai's CoCoSys Lab, are already benefiting. Johnson studies fluctuations in how mice perform the task, while Zang investigates how neuronal function changes as mice age. Johnson: 'The open science aspect of this project is unique. Because everything is documented in such detail, I can really understand and use the data. That's not usually the case when you take data from another lab for your own research.'

Sharing data

The IBL also aims to address a wide range of new neuroscientific questions and will continue to share tools, datasets, and platforms with the global research community. 'We would like diverse groups of scientists to join the IBL, run their own projects, and benefit from this open exchange of data and ideas,' says Tatiana Engel, associate professor at Princeton and IBL member.

Urai: 'It took enormous effort, funding, time, and expertise to gather these data, and we want them to be fully utilised. Now, any researcher with a strong question-even without their own mouse lab-can work with them. This makes neuroscience more open, accessible, and reproducible.'