How do alterations in the intricate machinery of the brain lead to autism, schizophrenia or depression? Today, large datasets collected through methods such as genetic testing and neuroimaging provide scientists with rich sources of information through which to decipher the biological roots of these types of complex conditions. Analyzing these data remain an immense challenge—but investigators at McGill University’s Ludmer Centre for Neuroinformatics and Mental Health are taking on this task with an interdisciplinary team and a combination of scientific tools.
“The Ludmer Centre’s principal thrust is to use brain imaging, genetics, epigenetics, psychological tests and various other kinds of neuroscience data to understand the machinery that underpin mental health disorders,” says Dr. Alan Evans, a professor of Neurology and Neurosurgery and Psychiatry and co-director of the Ludmer Centre. “It is focused on developing machineries and capabilities that can then be applied to many different scientific questions about normal brain function, development and brain disorder.”
The Ludmer Centre was established in 2013 thanks to a $7.5 million gift from Irving Ludmer, BEng’57, LLD, who, after graduating from an engineering physics program at McGill, developed an interest in psychiatry and mental health.
Today, the Ludmer has three scientific directors who each bring a unique set of expertise to the centre. Evans, who is the director of the McGill Centre for Integrative Neuroscience at The Neuro (Montreal Neurological Institute-Hospital), heads the neuroinformatics and neuroimaging components; Dr. Celia Greenwood, BSc’85, a James McGill professor and a senior investigator at the Lady Davis Institute for Medical Research at the Jewish General Hospital (JGH), oversees the statistical genetics and analysis aspects; and Dr. Patricia Pelufo Silveira, PGME’07, an associate professor and researcher at the Douglas Research Centre leads the work on gene-environment interactions and mental health.
In addition, the members of the Ludmer Centre come from various disciplines, including neuroscience, psychiatry, physics, mathematics, statistics, genetics and engineering. “We have a very interesting and eclectic community of people,” Evans says. “Keeping all of those people talking to each other is, in itself, an interesting challenge—oftentimes they have to work just to find the right vernacular to talk to each other because they are so wedded to their own disciplines.”
“Working in an interdisciplinary space, with colleagues from different backgrounds—and building all of that on strong, high-quality data—is a nice launching pad for really excellent research,” Greenwood says.
A new arsenal against brain disorders
Research at the Ludmer Centre has already led to important breakthroughs. In one recent study, Dr. John Lewis, a postdoctoral researcher in Evans’ group, examined the data on early-life brain connectivity collected with functional magnetic resonance imaging (fMRI) in more than 200 infants. Through this analysis, Lewis demonstrated that it was possible to pinpoint aberrant brain patterns in children who would later go on to develop autism at as early as six months of age.
“You would not make a diagnosis of autism at six months based on just brain imaging,” Evans says. “However, combining that information with behavioural, genetic and epigenetic measures may allow us to say that this child has a signature that is likely to ultimately manifest itself as autism down the road.”
Another ongoing project, led by Silveira, is aimed at integrating gene expression data to the calculation of polygenic risk scores, in a technique called “ePRS.” Using a combination of neuroimaging, genetics, and behavioural studies, Silveira’s team is exploring how the various combinations of genes, grouped in networks associated with biological function, are associated with changes in brain features and risk for psychopathology. “We have created a specific technique in the lab called expression-based polygenic scores,” Silveira says. “We use information from the function of the genes in the tissue or how much those genes are expressed in a specific brain region, for example, to calculate polygenic scores.”
In the past, researchers used to consider only single polymorphisms, despite the fact that the genome is so complex, explains Silveira. Nowadays, however, scientists have a better understanding of the multiple genetic factors that may underlie a disease. Even so, the focus is often on specific biological functions linked to a particular gene, although genes actually work together in networks. “That’s where the expression-based polygenic risk score comes in,” Silveira says. “What we do is use gene expression datasets to identify biologically relevant gene networks, and that’s what informs the development of the score.”
While the rapid growth of datasets provides an increasingly powerful resource for understanding the complex mechanisms underlying many brain disorders, they also make analysis a bigger challenge. “Understanding sound statistical principles is essential,” Greenwood says. To ensure that researchers are well informed, Greenwood has been working to facilitate training programs to enable researchers to learn how to carry out thorough assessments of these data.
At the same time, Greenwood’s team is also working on creating better tools, such as lack-of-fit statistics—a method to identify the correct model for a dataset by ruling out other models—for complex, high-dimensional datasets.
“We’re suddenly in an era where the sample size is becoming larger and larger really fast,” Greenwood says. “That’s where the Ludmer Centre comes in. Our focus is: How do we actually find those signals that are in these data? How do we model them? How do we understand them? How do we do it in a way that is high quality?”
Open science and collaboration
One of the key values of the Ludmer Centre is open science. “We’re very much committed to placing our data and software analytic tools into the public domain so that other people can analyze these data and come with fresh ideas to complement what we’re doing,” Evans says. “The general justification for open science is that if you have 100 scientists looking at the data, you’re going to get a lot more bang for your buck than if you have one scientist looking at it. In the long run, this acceleration in knowledge gathering will lead to a faster cure for specific brain disorders.”
The pipelines produced in Silveira’s ePRS study, for example, are being made publicly available through a pan-Canadian network of data-sharing called the Canadian Open Neuroscience Planform (CONP), led by Evans, which Ludmer Centre scientists helped establish.
Silveira says that one of the Centre’s goals in the near future is to consolidate the use of large datasets—such as the UK Biobank, which contains the genetic and health information from more than half a million participants in the United Kingdom, and Adolescent Brain Cognitive Development (ABCD), a repository of data generated from a large-scale study of brain development and health—and make the pipelines for processing the data more accessible to researchers.
The need for better access has been accelerated by COVID-19—as many scientists work from home, the need for remote access to data and computing capacity has risen dramatically, Silveira says. “Of course, the pandemic is terrible, but it has forced us to move forward with these initiatives—and that’s a good thing.”
This collaborative environment has enabled discoveries that extend beyond the research into mental health. Through analyses of large-scale datasets of gene expression and epigenetic information in brain development, Dr. Claudia Kleinman, an assistant professor in the Department of Human Genetics as well as an investigator at the Lady Davis Institute for Medical Research at the JGH, and her colleagues were able to identify the cell of origin of several types of aggressive pediatric brain cancers. This research made Québec Science magazine’s annual list of top 10 scientific discoveries in 2020.
The conversations that occurred within the interdisciplinary environment of the Ludmer Centre provided the crucial brainstorming that sparked the ideas that ultimately culminated in that study, Kleinman says. “This is a rich, stimulating environment for scientific discoveries. You never know what will come out from those discussions.”
Since 2013, the Centre has grown from its three original directors—Evans, Greenwood, and Dr. Michael Meaney, a professor and researcher at the Douglas Research Centre—to a thriving community of scientists, including 13 principal investigators and more than 100 researchers. “There is a plethora of really brilliant young scientists who have come here in recent years to study brain disorders and brain health,” Evans says. “It is so exciting to watch this new generation of data scientists and clinical researchers take centre stage. I could not have imagined this growth in the Centre when we first launched it less than a decade ago. We are forever indebted to Irving Ludmer’s vision and support.”
The way that the Ludmer Centre has grown into a hub of multidisciplinary research activity is “exactly what I dreamt of,” Irving Ludmer says. “I never thought it would be, in my lifetime, spread out the way it is. But it’s exactly what I wanted to see happen.”