September 2001

Under Biology's Hood

Continued from page 2

By Technology Review

TR: Why is the lay public's knowledge of medicine and science important?
Hood: In the end it is the public that places two limitations on how we do science. One is funding and the second is the regulations that govern how we do science. Accordingly, it is imperative that we somehow reach out to the lay public and educate them about these issues.

TR: In a nutshell, what is systems biology?
Hood: Suppose you wanted to figure out how a car works. The way biology has done it in the past is to create a group of specialists that would study individual parts-the wheels, the brakes and the ignition. Each group of specialists would talk within their own group but not to the members of the other groups. What systems biology does is attempt to use discovery science to define all the elements in the system, all the components of the car. Then it perturbs the car-accelerate, brake, etc.-and attempts to define the relationships of the elements one to another at various levels-mechanical, electrical, etc. Finally, systems approaches integrate these different kinds of measurements and data in a way that one can begin to formulate graphical displays and finally mathematical models that ultimately will give one insight into the structure and functioning of the car.

That is basically what systems biology is about. It is taking a biological system, identifying its elements, perturbing it in a model system, capturing information at the DNA, RNA, protein, protein interaction, informational pathway and informational network levels, and integrating and graphically displaying [that information], and then developing mathematical models that will describe the structure and behavior of the system.

TR: What does that allow you to do?
Hood: It allows you to understand the system and how it functions. If you are a pharmaceutical company, it suggests that there are critical points in these informational networks at which one can begin to attack the system, manipulate it, circumvent the limitations of defective genes. We are at the earliest stages of learning how to do mathematical formulations. Once we do, I think it will transform how one identifies drug targets, how one deals with potential side effects of drugs, and how one determines whether a particular drug that has already been approved might do something else that is even more interesting.

TR: Is all this modeling moving biology out of the wet lab and into the computer-what some call "biology in silico?"
Hood: Not at all. The message is that we have to integrate the computational tools with the data generated from biological tools. The systems-model process is really iterative with data generation, modeling, data quantitation, etc. The first time you go around the loop, you find that your model is not very good, so you have to do more experiments to improve it. This process repeats itself until you get to a place where the predictions you can make with the model are in conjunction with the experimental data you generate. You will never make progress in biology if you believe you can attack biological complexity solely in silico. The heart of biology is complexity, and we are going to unravel complexity only by doing biological experiments. But the integration with modeling and the graphic display of complexity is a central feature of what we're trying to do at the Institute for Systems Biology.

TR: Why did you start the institute?
Hood: I moved from Caltech to the University of Washington School of Medicine in 1992 to start the Department of Molecular Biotechnology. The vision of the department was to be cross disciplinary; that we were to hire engineers, applied physicists and computer scientists, as well as biologists. By 1995 or so, we had filled all the space we had been allocated. The department was enormously successful. We had terrific people. We had great funding. So I went to the president of the University of Washington and asked to build a new structure to expand the department in keeping with the already apparent opportunities of systems biology. The president said no, that there were eight or nine other projects in line and that I would have to wait 10 or more years before I could get a building. I then decided to start the institute. I spent about four years trying to create it within the School of Medicine at the University of Washington, but in the end, I had to resign and establish an independent nonprofit institute. When all was said and done, I realized that the university culture and bureaucracy just could not have sufficient flexibility for the needs of an institute attempting to respond to the opportunities emerging from the Human Genome Project. The Institute for Systems Biology has been operating for a little more than a year. We have seven cross-disciplinary faculty and have grown from a staff of two to a staff of 160. We've established six technology platforms, we've been successful in getting grants and industrial support, we've established a number of industrial partnerships, and we've published a series of exciting papers, including one that is a proof of principle for the Institute.