Please tell us about your research.

My research is in a field called bioinformatics. Bioinformatics is a field with the goal of developing new things, such as methods and algorithms used to analyze life science data using information science techniques. In particular, I am focusing on gene expression analysis, which is a method to analyze the expression levels of genes. Gene expression levels are table data, like data you would open as a spreadsheet, written in very large rows and columns. It is similar to table data where each gene is lined up in the rows and many samples are lined up in the columns. The research is to analyze similar patterns in such table data, or to analyze which factors should be paid attention to. At the same time, I am also working on cell image processing. The purpose of this research is to analyze images of cells taken with a microscope. Cell image data can tell us a lot when we look at it, but in order to analyze it by computer, we need to extract "variables". In the case of gene expression analysis, which I already mentioned, it is possible to analyze the data in rows and columns in the form of a spreadsheet from the beginning, but in the case of images, the starting point is to think about the expression of various features from what kind of image this is. Once the features are extracted, the process is similar for both cell image processing and gene expression analysis. In other words, the analysis is based on how to find similarities and separate differences.

Can you tell us about how you decided to study this field?

For many students, their first motive to start research is choosing a laboratory when they are in their fourth year of university. I was in my senior year during the years 1999-2000. It was the dawn of genomics, just after the human genome was elucidated, and genomics was attracting attention worldwide.

I have always been a fan of science fiction, and I think genomes were often used as a subject in science fiction at that time. There were many novels, movies, and games based on the genome, such as "Parasite Eve."
Another thing that marked the time was IT. I entered university in 1997, so I was in high school when computers started to become popular, such as with Windows 95 being released. At that time, I thought that if I was going to study at university, it would be IT, and if I was going to do research, it would be IT + genomics. With this background, I entered the Seminar of Genome Information Engineering.

When I was assigned to the Seminar of Genome Information Engineering in the Matsuda Lab as a fourth-year undergraduate student, gene expression analysis was just beginning to appear as a research topic. So, the first research I started as a student was expression analysis. You may be familiar with the kind of image analysis called "heat map", which has a clustering tree structure on the top and side of the image which can be rearranged and organized at will. It was the first time that such an analysis and visualization method was used with microarrays. I thought that the Eisen plot, with its beautiful red and green patterns, was beautiful, and that was one of the reasons I started my research. From there, I continued to research clustering of expression profiling during six years (B4 to D3) as a student at Matsuda Lab.

At that time, I had no idea that I would make this my lifelong research. However, I would like to ask the students - do you ever feel that there are still too many things you don't understand about what you studied at university, even though you are studying the most advanced things? When you think things like, "After graduating, it's just work," "No more studying and research," you may still feel like there is so much left that you don't understand. That's how it was for me when I was a master's student. I studied hard at university and learned more, but there were too many things I didn't understand. Above all, I thought that once I started working, I would remain clueless about it. So, I went on to become a doctor, and by a stroke of luck, I am continuing my research at the Matsuda Lab.

When did you start your research on image analysis?

It was about 10 years ago that I was approached by Dr. Matsuda to do research on image analysis after I had exchanged ideas with a professor from the School of Medicine at a seminar. It was a time when research on expression analysis was at a standstill, and I felt that image analysis would be interesting, so I decided to start studying it as it was a new field. Nowadays, image-related research is very stimulating because there are various research subjects such as deep learning, all of which are being studied in depth. However, because of this, there are so many people researching it, so it's hard to make up my mind to specialize in image analysis and work hard at it alone. I am working on it in a shallow and broad way, leaving an escape route.