UNB Research

“The working of the universe is beautiful, and, well, you need to smile about that sometimes. You need to appreciate it.”

Dr. Stijn De Baerdemacker is the Canada Research Chair (CRC) in Theoretical Chemistry at the University of New Brunswick (UNB). He’s also an associate professor of chemistry and associate research director of UNB’s Research Institute in Data Science and Artificial Intelligence. Beyond that, he’s a physicist, a mathematician, an artist, something of a philosopher and still more.

We’re tempted to call him a renaissance man.

Where some see complicated equations, confusing technologies or ordinary words, De Baerdemacker sees patterns, beauty, universality and connections that may hold the key to solving our most complex questions and our most challenging issues.

“When we look at the proteins associated with Alzheimer's disease, for instance, we find that these long strands of molecules have curled up in specific ways,” he said. “The way they’re folded is very important for their biological function, and if they fold up in the wrong way, they curl into spikes and pierce through the cell membrane. Obviously, this isn’t good.”

“These proteins, and the ways they fold, are very complex, but they can be represented mathematically and abstractly in a way that lets us explore how and why they fold the way they do—and, perhaps, find ways to stop them.”

With a research portfolio that includes quantum chemistry, quantum computing, machine learning, and a lot of math, we sat down with De Baerdemacker to learn more about what he does, how he does it, and why it matters.

[ed.: If you're curious about how quantum computing and machine learning and AI are relevant to chemistry, good news: So were we! Dr. De Baerdemacker was kind enough to walk us through some of the key concepts and their application to his work at the links.]

Tell us a bit about yourself and your background.

I am an associate professor in the department of chemistry and the Canada Research Chair in Theoretical Chemistry at UNB.

I completed my undergraduate and master’s degrees at Ghent University, in Belgium. I began a joint program in mathematics and physics and eventually completed a program in theoretical physics. I think that combination of math and physics has, in a way, shaped me along with the rest of my career.

Originally, I wanted to be an engineer, but in order to know how systems work, I felt I needed to know what the physics underlying them were; and in order to understand the physics of those systems, I needed to know the math behind the physics. So, I went to the root, and I've been slowly climbing up the ladder, one step at a time—from the math that underlies so much—maybe all—of our world, making a positive impact by solving big, real questions and challenges.

I did my master’s work in nuclear physics and followed that with a PhD in theoretical nuclear physics, furthering that interest in the mathematical structures underlying this field.

Our dean, Dr. Sanjeev Seahra, once said that I am looking for excuses to do math. But through it all, I’ve always wanted to keep my work in touch with real situations and applications.

Following that, thanks to a very generous travel grant I received from the Flemish government, I was able to travel the world and explore my future.

I eventually ended up in the mathematical physics group at the University of Toronto, trying to determine what interested me and where I was going to be able to make a positive impact.

I started looking into topics like how many-body systems can be solved in efficient ways using symmetries, using math.

I then went to the United States and to further diversify my explorations. I started looking into quantum computing—this was still very early on in that development. My first conference presentation on quantum computing was in 2010, about 10 years before the first prototype was on the market.

I also started exploring quantum chemistry, and met Dr. Paul Ayers, who was the CRC in Theoretical Chemistry and Chemical Biology at the time at McMaster University. This marked what is probably the most significant turning point in my career, as it appeared that the mathematical structures I had been exploring in Toronto also turned out to have important applications in chemistry. I became a chemist from that point onwards.

Eventually, after a few more postdoctoral fellowships and similar adventures, including joining the multidisciplinary Centre for Molecular Modeling in Belgium, I ended up here at UNB.

I was really glad to end up at UNB, which I have found big enough to support a wide range of research, while still small enough to have a more personal touch and close connections with colleagues and students.

Tell us about your chair. Does it build on these experiences?

My research program definitely reflects these experiences. Currently, I’m working on three major pillars with my research group.

The main question we’re exploring is whether we can develop computational and mathematical methods that can be implemented in computers, which will then help us to compute the properties of molecules in a better way.

The other two pillars we are looking into are emerging technologies to use in creating these new methods: machine learning and quantum computing.

With machine learning, we're especially looking into the question, why are the machine learning models so successful? It's baffling how well these approaches work.

Similarly, quantum computing is a bit like a device that was given to us by some alien society. We’re really trying to see how much we can gain from this new technology.

What do you hope will come out of this research? Why does it matter?

I think our work is extremely important. So many of the issues that we face in the world involve some sort of chemistry: Technological challenges such as, how do we make new materials? Medical topics like how do we find new medications to help people? Agricultural pressures that force us to consider, just how do we feed people? There’s always a chemical involved.

Now, the biggest issue we face in chemistry, here, is that the diversity of possible chemicals is enormous. It’s mindbogglingly huge.

The number of different drug compounds we could make is somewhere in the order of 10 to the power of 60. That’s a one followed by 60 zeros. It’s in a space where the number doesn’t even make sense anymore. To put this number in perspective, there are only 10 to the power of 80 protons in the entire universe.

We hardly have enough matter in the universe to make each possible drug compound, let alone the time we would need to test and evaluate them.

[ed.: To expand on the above, if there are 10^60 possible compounds, and only 10^80 protons in the universe, we could roughly generalize as having enough matter to make, at most, 10^20 molecules of each substance. A single ibuprofen tablet weighing 500mg has about ten times that number of molecules.]

We need to be able to short circuit this whole discovery process that presently takes decades and millions, if not billions, of dollars to bring a new pharmaceutical product to market. We need to be able to accelerate that by computing and simulating things where we can break through our current computing walls and get the result in a few days, not in a few weeks or a few months.

Similarly, understanding the protein folding process involved in Alzheimer’s, or being able to understand and compute particular properties at the quantum level as electronics continue to shrink, would have an enormous and transformative impact on society.

But these are huge goals, of course, even if you can summarize them in a short sentence.

We call this moonshot research. We know it’s a big challenge. It's ambitious, but if it’s feasible, if we succeed, it will be a giant leap for humanity.

Of course, these are massive questions that will take the minds, time and efforts of a great many scholars. It might be achievable, if I’m optimistic, by the time I retire.

So, you have some pretty ambitious goals for the long term. What do you want to accomplish in the shorter term, during your tenure as Chair?

I want to help us understand how neural networks and machine learning come up with the conclusions they do: Why do they come to such good conclusions, and how we can use them to solve chemistry questions?

Similarly, I want to better understand how quantum computers do what they do, and how they can help us tackle problems that require us to juggle multiple possibilities along the way.

If I can accomplish this, I think we will be putting ourselves, as a research community, in a good place to continue our journey toward these greater goals.

More information

Dr. Stijn De Baerdemacker [UNB profile | CRC profile]| Department of Chemistry | Faculty of Science (Fredericton)

Research at UNB | Graduate Studies at UNB | Postdoctoral fellowships