A little bit about you:

I am a Bayesian statistician and data-driven scientist, focusing on developing new machine learning methods. I am excited to find a new pattern emerging from big biomedical data. But I become more excited if I can design and implement a machine learning algorithm that can make such scientific discoveries in a massive scale.

What would people be surprised to learn about you?

Perhaps, it is not about me and not surprising. It was when my research interest in deep learning methods was so high, trying to understand the mechanisms of Alzheimer's disease. I went to my daughter's daycare at MIT, and one teacher showed me a picture drawn by my daughter. It was a shape of brain coloured yellow/brown. I asked her teacher, "is it a human brain?" Then, she said, "Well, I asked students to draw his/her favourite body parts, and your daughter drew this and named its neural networks." I must have talked a lot about my research to my three-year-old daughter. Besides, my wife studies brain science.

What does a typical day look like for you and what are you currently working on?

I spend most of time looking at Emacs (a text editor) or shell screens, writing codes and analysis scripts of genomics data. These days I am developing a suite that enables us to analyze millions of single cell data efficiently to identify known and novel cell states.

What are you most looking forward to in your new role with BCCA/Dept of Pathology?

Although I am a bit introvert, I am most excited to meet new people, especially students and faculty, with whom I will work closely. I think of the discipline of pathology as a pursuit after causality, just like a detective would do for a criminal investigation. It demands the work of a multidisciplinary team. No one has a perfect clue, but we would collectively have a complete picture of causal mechanisms of cancer and other related diseases.

What did you learn as a researcher the hard way?

In computational biology research, finding a meaningful problem and defining the problem in a precise model is perhaps one of the most challenging tasks. When new trends of technology emerge, out of excitement, we tend to quickly adopt new technology to our problem without rationally thinking about its feasibility. I have made such a mistake a lot. I almost believe that it is better to design algorithms and implement tools in a problem-specific manner.

What do you like most about your job?

I am a little obsessed with details and visualization. To me, it often takes quite a time until I see the fruition of satisfying results and figures. Still, nothing is more satisfying than staring at these products. I also like the fact that researchers are constantly teaching each other regardless of positions. I enjoy learning new things from colleagues and students. It is never a one-way street.

What is the most challenging part of your job?

My job is to solve real-world biological problems using computational and statistical methods in the contexts of research and mentoring. I should be bridging between experts in two completely different fields. I need to convince a pure theoretician to work on a biology problem and teach quantitative backgrounds to biology-oriented researchers. Both are equally challenging.

What do you believe has been the greatest advancement for medicine in the last 50 years?

Human genome sequencing. With the help of sequencing technology, I believe we will see even more rapid advancement than ever seen before.

Going into the future, how do you believe medicine/science may change?

Medicine is an evidence-based science. As we collect more data, both actively and passively, it is not difficult to imagine that the very definition of many diseases will change. Therefore, the procedure and basis for diagnosis in many medical problems will be revised. We will be able to improve the accuracy of diagnosis by a collective and data-driven decision-making procedure, which involves multiple clinicians across the globe, including machine-learning algorithms. However, I do not share a view that Strong Artificial Intelligence can autonomously make more precise decisions than human clinicians. Humans are much better at reasoning causal relationships with only a few random examples.

What is the most helpful advice you’ve received?

I heard this from my postdoc advisor: Some computational problems are so easy to biologists and medical doctors. (Therefore, do not try to solve everything computationally)

What publications do you regularly read?

I follow great minds in machine learning and science through Google scholar.

What are some causes you care about?

I care about open creativity (if I understood the question correctly). I would like to keep on expressing creativity in research and attract more people to computational biology, thinking, "Well, I can do better than this guy!

How do you like to recharge?

I enjoy cooking although it does not mean that I am a good cook.

What are some words of wisdom or perhaps inspiration you wish to share with your colleagues?

Where is the finesse?
-Gordon Ramsey