I graduated in 1995 with a Ph.D. in theoretical physics. At that time, many of us felt that demand for physicists was at its all time low as funding from the Canadian government continued to decline. So I turned my attention to other areas of research where I could apply my skills. My interest in computer applications in biology was really due to my wife who happens to be a molecular geneticist. She told me that the biology community could use many mathematicians and physicists with excellent computing and analytical skills. So in 1996, I got my first bioinformatics position. I was given a Medical Research Council (Canada) research fellowship, looking for efficient ways to determine protein structures using data collected from Nuclear Magnetic Resonance (NMR) experiments. Since then, I have been a bioinformatics specialist, now with the Organelle Genome Megasequencing Program (OGMP) in Montreal.

There are two reasons for me to keep working in bioinformatics. The first, and most important, is scientific curiosity: Like many people, I was intrigued by the fact that the blueprint of life is encoded by sequences of a four-letter alphabet and, regardless of this seemingly simplistic nature, there are still no definite ways to decipher these genetic messages. This basic simplicity in coding to create complex messages is intriguing, and enough of a puzzle to keep the mathematical inclinations in my personality stimulated. When this is combined with the concept that the long, two-dimensional strings of nucleic or amino acids that are DNA and protein actually exist in three-dimensional space and interact with one another in accordance to the laws of physics, the scientific problems and issues are so complex that you could very happily spend an entire lifetime trying to understand them. There is definitely a lot just waiting to be discovered.

The second, more practical reason for my career choice in bioinformatics is the better and more numerous career choices the field offers. At the time I started in this field, there were fantastic employment opportunities in bioinformatics for physicists with good computing skills. Those opportunities have, if anything, increased in scope and number since then, and, because many of these opportunities are in companies, bioinformatics becomes a perfect career transition from the academia physics bench to the world of industry.

A "Day in the Life"

I keep very regular working hours. Like many of us, I work Monday to Friday, from 9 a.m. to 5 p.m. I typically begin my day with a system check, looking for trouble spots and signs of network intrusion. The rest of the day would most likely be filled with heavy programming or debugging sessions. I keep regular communication with my colleagues and our in-house biologists to ensure programs under development or modification are proceeding at good speed and in accordance with specifications.

But there's a whole lot more to the job than heavy programming. Most people might think the life of a bioinformatician is about sitting at a computer terminal, writing programs. Although this is a big part of the job, it is not the whole job. What I find most interesting about my job is that it is not about just writing programs. I often have to apply techniques and principles from other branches of science to the biological problem at hand. This interdisciplinary work is fascinating, challenging, and rewarding.

Skill Sets Required to Be a Successful Bioinformatician

There are really two sets of skills required to do my job. These are the "technical" skills I need to actually do the programming, and the "scientific communication" skills I need to design the software.

On the technical side, there's a lot of traditional programming knowledge required. A large amount of data analysis and software development is being carried out on the UNIX platforms. So a strong background in UNIX is essential. For software development and process automation, I use C, Bash, Perl, Python, and Java. I am also the system administrator at OGMP. So, in addition to good trouble-shooting skills, a good understanding of networking (TCP/IP) protocol and network security, abilities to manage user accounts, perform backups, and manage network information service (NIS) are all very important. I am also providing technical support for our GOBASE project, which maintains a sophisticated organelle database with a WWW front-end. That requires installation and configuring our Sybase database server, basic SQL, HTML, and CGI programming skills.

The second set of skills are more "biological sciences" in nature. Because I need to communicate with molecular biologists, designing or modifying computer programs to accommodate their needs, understanding of basic biological concepts such as DNA, gene, exon, intron, open reading frame, and their typical arrangement within a genome is very important. Also, because bioinformatics is such a multidisciplinary field of research, abilities to apply techniques available from other scientific areas (such as mathematics, statistics, physics, and biochemistry) are a necessary part of my job.

The Future of Bioinformatics

As the field expands, and focus shifts to inferring 3D structures based on primary sequences and data obtained from other types of structural experiments, I see tremendous opportunities for young scientists with various expertise in the physical sciences. I see opportunities in three separate but overlapping fields: comparative genomics (sequence analysis), protein modeling (x-ray crystallography and NMR spectroscopy), and computational proteomics (protein folding, conformational sampling, docking, and molecular dynamics simulations). I believe these are the "hot" fields that will lead to plenty of employment opportunities for biomathematicians and statisticians, computer scientists, chemists, computational biochemists, and physicists. Right now, the focus is on comparative genomics. But as the research focus begins to shift, new opportunities will surface.