Harvard’s Robert Lue, one of the ‘directors’ of a popular animation of cellular activities, talks about how the computer animations of today can become the simulation models of tomorrow.

Corie Lok

In biology classrooms and cell and molecular biology labs around the world, students and scientists have been awed by “The Inner Life of the Cell”, an eight-minute computer animation showing biochemical activities inside the cell. More than just flat shapes moving across the screen, the video depicts in vibrant colors and dramatic, three-dimensional detail proteins binding to receptors, the assembly and disassembly of protein filaments, and the transport of molecules through the cell—activities that biologists until recently could only picture in their minds.

The movie was produced by a medical animation company in Connecticut for the Biovisions project at Harvard, which was founded by Robert Lue, director of life sciences education at Harvard and a professor in the molecular and cellular biology department.

Since the release of the three-minute version in the summer of 2006 and the eight-minute version in early 2007, the animation has been used as a teaching tool in high school and college biology classes across the country. Scientists have used it in their talks at meetings. Dozens of researchers have asked Lue and his colleagues to collaborate on animations of the cellular systems they study. Lue’s group is continuing to produce more movies focused on other biological processes such as embryo development. Lue recently spoke with Nature Network Boston about what these kinds of visualization tools can do for science and science education.

What do you hope to achieve with these animations?

A lot of colleges, including Harvard College, have been rethinking how to teach undergraduate science. Integrating concepts across the biological and chemical sciences is a high priority. But we’ve found that undergraduate students have a difficult time tying together concepts they learn in a molecular biology course. So we’re trying to provide integrated, synthesized views of biological processes and that’s where the animations have been very effective. It allows students to see things and to see something is to begin to understand it.

For the last three years, we’ve been assessing how the use of coherent animations changes the way students are able to apply the knowledge learned in class. There seems to be a clear indication that these animations do make a difference in terms of the students’ ability to interpret experimental data in a way that shows an understanding of connected issues. So far, we’ve assessed about 600 students at Harvard.

The other crucial piece is engagement. I cannot begin to tell you the response we’ve gotten from high schools and grade schools. Many high school teachers have told us that the animation captures their students’ attention enormously and makes them want to understand what those things in the animation are.

All of our animations have a cinematic quality: there’s angle and pacing, etc. Part of that is to pull viewers in. In my view, that’s not something negative. Our task is to engage the viewer. But the cinema never alters the solidity of the science. It’s the combination of cinema and science that’s very important.

But the videos aren’t meant to be completely realistic simulations, right?

You don’t want to simulate all aspects of reality, because reality is, in some ways, far too complex. If we showed the high density and full motion of the macromolecules and events in the cell, you wouldn’t be able to see specific events.

So what we do is in between a diagram that you would see in a review paper and a high-end simulation that attempts to show every physical aspect. Our animations bring together key pieces of scientific data with the goal of communicating a specific point. If something is not relevant to that point, one can decide to not show it.

We’re also working on detailed annotations that will be presented with each animation and will explain what we’re showing and why we chose to show one thing and not something else. It will link up to the actual science: the multitude of papers used in the making of the animation. I think this is necessary for these animations.

Have there been any surprising uses of the videos?

We were approached by a major studio in Hollywood to use the animation in a science fiction film that’s coming out this year. The difficulty with that use is that when you take an animation that’s based on real science and place it in the context of a fictional film, the risk is that it gives scientific validity to a work of science fiction. We felt that that combination of Hollywood with what we’re doing would not work. So we ultimately said no. This was especially disappointing because I am a huge fan of the directors of this film.

Most of the uses of the animation so far have been for education and communication. With the rise of systems biology and modeling of cellular activities, what role do you think animation will have in furthering research?

In many ways, systems biology and high-end visualization are meant for each other; systems biology seeks to tie everything together into a more coherent model of the whole and that is what this kind of visualization also seeks to do. So I expect these two things will come together seamlessly.

You can think of these animations as the visual front end. What we want to do is make sure that the back end—the hard-core computational simulations of events—can be easily connected to this front end. We may simplify the movement of molecules, for example, but a scientist could plug in his or her algorithms to make the molecules move the way they do in reality. Then you can think about this as a research tool.

Where do you see biological animation going in the future?

We’re trying to establish a standard visual language, so that the same things in the cell are rendered in the same way from one animation to the next. Down the road, if we develop this visual language appropriately, there’s an enormous potential for the development of software suites that would allow scientists to model far more easily. Right now, to do these animations, sophisticated animators are involved, but I would love to make some form of this available for anyone who’s relatively computer-savvy to do it on their own.

It could be very powerful. Think of it as a super-modeling environment. There are two ways this could go: allow an individual researcher to do it at his or her desktop, which I think is ultimately what we want. The other is to have centers for science visualization. So when a scientist has an important model, he or she can work with a center to visualize that model. That may actually be an interim point.

I believe this will be available for anyone to do on their desktop. There are things we can do with available animation programs today that would have been unimaginable a few years ago. Given what has happened in the last five to 10 years in animation, I think that in the next five to 10 years, the power of what we can do on the desktop will be extraordinary.