Faculty Spotlight on Dr. Andrew Lim
By: Jabir Mohamed
Dr. Andrew Lim, MD, FRCPC
Scientist, Evaluative Clinical Sciences, Hurvitz Brain Sciences Research Program, Sunnybrook Research Institute
Staff Neurologist, Department of Medicine, Sunnybrook Health Sciences Centre
Assistant Professor, Department of Medicine, University of Toronto
Associate Member, Institute of Medical Science
Dr. Andrew Lim is perhaps best known for discovering a genotype associated with earlier wakefulness. The clinician-scientist at Sunnybrook Hospital is investigating the connection between genetics in sleep and circadian biology, using a combination of mathematical models and statistical genetic tools. The IMS Magazine sat down with Dr. Lim to learn how his interest in sleep medicine evolved, and the implications of his current work.
Could you begin by talking a little about your personal background, where you grew up and your early education?
I completed my undergrad at the University of British Columbia, where I majored in microbiology and immunology. I then came to Toronto for medical school, and that was the beginning of my eventual affair with neuroscience. My interest in sleep medicine developed during my neurology residency, through the work I did with Drs. Brian Murray, Richard Wennberg, and Andres Lozano. Thereafter, I ended up at Harvard for more clinical and research training.
What was your thesis on and how did you decide on it?
The thesis evolved in an odd way. Clif [Saper], my supervisor, had an ongoing collaboration with some folks in Chicago who had postmortem brain tissue and actigraphy data from Alzheimer’s patients. Although the actigraphy data was being collected for other reasons—specifically, to measure physical activity—Cliff wondered if we could obtain some useful information about sleep architecture and its relationship to sleep circuitry.
The original project centered on the ventrolateral preoptic nucleus, and, as you know in rodents, lesions of this region lead to a deficit in sleep continuity. However, it wasn’t clear what the best actigraphic correlate of sleep fragmentation was going to be in these patients; after all, we didn’t have any EEG data. With the help of some folks at the Rey Laboratory, we developed a mathematical model.
In the end, the thesis ended up being three papers. There was a paper on the development of the mathematical technique, one on the correlation between the actigraphy and cell counts, and there was a paper correlating the actigraphy to cognitive outcomes.
What was it like to work with Clif?
Well, it was a wealth of opportunity. Clif runs a very productive lab, and he was an absolutely fantastic mentor, not just from a scientific perspective but also, from a career development and career advice perspective. His group is extremely well funded, so if ever something was needed, it was ordered. If you needed computing resources, you got it. If you needed mentorship in x, y, and z, you got it. People are very receptive to an email from Clif—it kind of solved your problems.
What got you interested in circadian biology?
It was very roundabout, but there are three things. The first thing was the seminar series on sleep biology and circadian biology I regularly attended as a graduate student. The other thing was the coursework—for various reasons, it was heavy on statistical and human genetics. Third was meeting Philip De Jager, a neurologist who also happens to be an expert in statistical genetics. David Bennett, the person who runs the cohort in Chicago, started a collaboration with Phil to explore the genome of the Alzheimer’s cohort and the thought sort of came as to whether we could take the actigraphy data and look for genetic influences. So under David and Phil’s guidance, I developed an algorithm to extract information about circadian rhythmicity from the actigraphy and genotype data.
Could you talk about the work your group does now?
There are three main things we do. One of the projects we have funded asks the question, what are common gene variants associated with various sleep and circadian phenotypes? Our target is a recruitment of 4,000 middle-aged folks in Ontario, 1,000 folks in their 20’s and 30’s from De Jagger’s group, and a collection of elderly people from the Chicago group. The plan is to genotype all of them and get actigraphy data, with the hope of identifying genes associated with sleep and circadian biology.
The other part is taking a look at circadian rhythms of gene expression and epigenomic modification, using mathematical modeling. We also have another project applying machine learning approaches to the analysis of human EEG data. The goal is to automate, to the greatest extent possible, the capacity to do large-scale analysis of human EEG data for research and clinical purposes.
Where do you see the field going in the next few years?
I think it’s difficult to talk about one direction that circadian biology is going. Certainly one thing is the idea of personalized, circadian biology-driven medicine—that through genetics, we could use information on individual variations in sleep and circadian traits to inform decisions from a clinical and occupational health perspective. There are also benefits from a chronotherapeutic perspective; if you know that drug X targets channel Y, but that channel Y is expressed preferentially at certain times of the day, then it makes sense to administer drug X when channel Y is expressed rather than when it’s down-regulated. This is particularly important when you have medications with short half-lives.
The other big thing is to understand how circadian biology influences risk for diseases. If you take a look at something like stroke risk, you’ll see a peak of stroke risk at a certain time, and a trough of stroke risk at another. If you were to take this profile of stroke risk and figure out what is causing this peak at a certain time and beat it down, so that the entire day you just had your baseline stroke risk, you could have a substantial impact on overall stroke risk. The change between peak and trough is by definition dynamic and temporary, so it may be a much better target for intervening then what causes the baseline level.
Do you have any advice to share for newcomers in the field?
Find out the great people who are doing exciting work in the field and who are going to teach you good techniques. You want good scientists who will teach you useful things, and to some degree, open doors for you from a mentorship perspective. These are all important things to look for when you’re making a decision on where to go.
I think this is true for two reasons. You can’t actually know where your career is going to go and it’s very difficult to plan. So, it’s good to be in a place where you’re surrounded by helpful people; it makes random good things more likely to happen, and when the random good things happen, it makes it much easier to take advantage of them.