A researcher at the Montreal Neurological Institute-Hospital (The Neuro), Evans is a James ³ÉÈËVRÊÓƵ Professor of Neurology and Neurosurgery, Psychiatry and Biomedical Engineering. He is also the Scientific Director of Healthy Brains for Healthy Lives and of the Ludmer Centre for Neuroinformatics & Mental Health.
Your lab, the ³ÉÈËVRÊÓƵ Centre for Integrative Neuroscience (), has developed many new neuroinformatics technologies—the tools that make big data analytics in neuroscience possible. What set you on this path?
As a child, I remember reading about the mysteries of memory in the magazine Scientific American. From that point on, I had an abiding fascination with the brain. My university degrees in the United Kingdom were in physics, medical physics and protein crystallography—moving me toward a synthesis of physics, mathematics and biology.
When the opportunity arose to become a positron emission tomography (PET) physicist at Atomic Energy of Canada (AECL), working with scientists at the Montreal Neurological Institute-Hospital (The Neuro) on a brain PET scanner, I jumped at the chance. After five years at AECL, I moved to The Neuro in 1984. It was the best move I ever made.
Neuroimaging is a relatively new field. Can you describe the paradigm shift brought about by advances in technology like MRI and PET?
In the 20th century, studies using imaging technology tracked which brain areas ‘lit up’ during an experiment. However, in the past decade or so, neuroimaging has embraced the notion of "connectomics"—the study of a structural or functional connection between brain regions. Connectivity changes constantly within the brain and is a more meaningful way to understand brain dynamics, over short and long periods.
Connectomics can be applied to the study of brain development and ageing, neuroplasticity, cognition and brain disorders—opening up an entirely new window into our understanding of the brain. Now, we can study how a child’s brain connectivity is sculpted as they learn to walk and talk. We are also able to map abnormal brain wiring that is associated with developmental disorders like autism. At the other end of the lifespan, we are better able to understand the mechanisms that underlie neurodegenerative disorders like Alzheimer’s and Parkinson’s. These advances make it possible to design therapeutic interventions, which target the specific mechanisms that lead to disease.
In your opinion, are we adequately training the next generation of neuroscientists to work with big data?
I believe so. Neuroscience is increasingly an interdisciplinary field, with trainees coming from a wide range of undergraduate backgrounds.
Indeed, the younger generations are spearheading the adoption of open neuroscience. Labs around the world are making their research data freely available to the scientific community for wider analysis.
For our part, we made the dataset freely available in 2013 and ³ÉÈËVRÊÓƵ has released a unique data set from subjects at risk of Alzheimer’s disease. The Neuro has adopted open neuroscience principles for all its research. Much of the energy for this revolution is coming from young trainees who are comfortable with the technologies and practices of data sharing and openness.
The challenge remains, however, of how to build a career path for these individuals at the faculty level. Too often, they are seen as neither fish nor fowl; not a good fit with traditional departmental priorities. Universities need transdisciplinary centres that can become a natural home for such individuals.
³ÉÈËVRÊÓƵ is at the forefront of neuroinformatics research in Canada and internationally. What does the future hold?
Neuroinformatics, while essential to the future of neuroscience, does not answer biological questions on its own. However, when combined with computer modelling, neuroinformatics can provide the methodology that allows these questions to be answered and creates an environment where new questions can be asked.
For instance, the complex relationships between an individual’s observed behaviour, what we can see in scans and what we can learn from their genes, can only be fully addressed through the application of big data analytics. This is the central theme of ³ÉÈËVRÊÓƵ's Healthy Brains for Healthy Lives initiative—a broad program of brain research that uses computational models to explore normal brain organization and brain disorders, specifically neurodegenerative, psychiatric and neurodevelopmental disorders.
This is where we see the next big leap in neuroscience research. Multi-modal approaches (e.g. merging imaging, genetics and behaviour), combined with the principles of open neuroscience, allow us to make greater use of artificial intelligence and machine learning to accelerate the pace of discovery, creating new opportunities for personalized medicine and quicker access to new drugs and treatments for patients. I think we have good reason to be optimistic about the future.