Complexity and Catastrophe
JB: Let's talk about the questions you're asking yourself.
MADDOX: There's an extremely interesting question that seems to me to be very urgent: how on earth is science going to cope with the accumulation of data, on a huge scale, of recent years. This relates to another question that hasn't been given enough attention in recent years: when on earth is biology going to become a quantitative science, like physics and chemistry - when there's good evidence to believe that it can't make progress in some fields without becoming much more quantitative.
The most successful efforts have come from the study of fear. Fear is a relatively tractable emotion, unlike love or hope which are difficult to pin down. It's always easier to study brain functions that involve clearly defined stimuli and responses than those that don't. For fear, you can easily create experimental situations where the onset of a simple stimulus that warns of impending danger elicits a set of stereotyped responses in an animal, like a rat, that are very similar to the kinds of responses that occur in a human facing danger. By following the flow of the stimulus through the brain from the stimulus processing pathways to the response control networks, it's possible to identify the basic neural circuits involved. We've done this for fear.
Let me give an illustration of what I mean. In a field like cell biology, everyone now has a clear picture of how the cell cycle is driven. You have proteins called cyclins, which are meant to interact with another protein called Cdk. Cdk plus cyclin activates two successive steps essential to the cell cycle. One of them is the replication of the cell's DNA; the other is the actual fission of the cell into two daughter cells. It seems to me high time that people recognize that the complexity of this system is so great that it can't really be dealt with in the simple way in which textbooks ordinarily deal with description, i.e. the explanation of events.
If, for example, you have an ordinary bacterial cell going through the process of cell division, it may be prompted to do that by some external signal in the environment; it may be prompted to do it simply because time has passed - twenty minutes is the length of time it takes the E. Coli to divide into two, and maybe just twenty minutes is up. There are several different molecular influences acting on this complex of cyclin and Cdk which is what actually triggers off the cell division. The complexity of the problem is so great that you can't comprehend it in the language I've been using; you can't comprehend it in the language of the textbooks, because it has become a mathematical problem. Nevertheless, very few people take this seriously.
There's more to it than that. I'm advocating that in the case of the cell cycle this is a specific biological problem. How do we understand the cell cycle? What makes a cell divide? What can we say about the competing influences on a cell - the external environment, the internal need of the cell, the need of some other cell in the same organism. How do these competing influences conspire to decide that the cell is now going to divide into two. What we need are mathematical models for saying what actually goes on.
There are other fields like that. Take the way in which the muscles in our arms work. Any molecular biologist will now tell you this understanding is one of the big triumphs of the past ten or fifteen years, that muscle fibers are made of actin and myosin, two proteins - and the idea is that the myosin molecule which is smaller than the actin molecule, acts as a kind of enzyme at the head of the actin fiber, that can ratchet itself along a parallel actin fiber.
Molecular biologists say, ah, we now understand how muscles contract. But nobody has done the thermodynamics of this problem. It's obviously a matter of great interest to the world at large to know how much energy is used. The molecular biologist will tell you it comes from ATP. ATP, adenosine triphosphate, is actually the universal source of energy in living cells, and everyone says that's fine, but actually what are the thermodynamic aspects? This question is not considered, and it's a crucial question, because that's the kind of consideration that would tell us when it is that muscles become tired, when they no longer function, or when they become rigid, and go into spasm. There are all kinds of important abnormalities in muscle behavior that would be explained by thermodynamics, if people put their minds to the task. The molecular biologists may have answered the "how ? question, but they will not be able to answer the "why?" question until somebody has done the thermodynamics.
What I'm saying underneath all this is that perhaps molecular biology got itself into the condition in which it's far too easy to get data, and therefore there is no incentive to sit down and think about the data and what they mean. But I'm sure that as the years go by, and not many years, people are going to have to be thinking much harder about how they get accurate quantitative data about the behavior of cells, muscles - all these things in living creatures.
JB: What is the relation of the acquisition of such data to the development of technological tools that allow you to formulate models and execute on those models. Are your perceptions related to the development in increased computational power?
MADDOX: The case I'm making actually doesn't depend on the improvement of computer technology, but what you say is absolutely right; that to solve some of these problems is going to require unprecedented computer technology. But let me illustrate it this way: suppose you want to understand how a cloud functions, a cloud in the sky. Sometimes you get rain out of a cloud, but not always - you see clouds up there but no rain coming out of them. Why is that? The reason is that in a cloud you have a constant upward and downward flow of drops of water, particles of ice and so on, and it's in a dynamic situation. For every cloud the bottom is at some temperature and the top is at another temperature - a lower temperature, of course. Sometimes this dynamic stability of the cloud becomes unstable, perhaps because there's a shift of the temperature, perhaps because something goes through it like a projectile, an aircraft perhaps which may leave a trail of condensation behind it if it's travelling through the humid atmosphere, But you can only understand cloud behavior and answer the question when will this cloud produce rain, if you have a model which can be in that case quite a simple model.
In the case of the cell, and the cell cycle, it's a much more complicated model I'm looking for, and there in reality one would need supercomputers to handle the model The models people have built so far - I'm thinking of John Tyson at the Virginia Polytechnic and Albert Goldbeter at the Free University in Brussels -have been handled on ordinary desktops, but everyone agrees they're not sufficiently refined. Once you start refining them you get into real problems. But there's some parts of science that can only be understood when you make a model. The cell cycle is one, the muscle is another, and each of them, being biological problems, are very complicated.