Researchers mutate digital organisms

PASADENA-In a study that could point to a new way of predicting what extraterrestrial life might be like, a team of California Institute of Technology, UCLA and Michigan State researchers have shown that "digital organisms" respond to mutations in ways closely resembling the mutations of actual organisms like bacteria, fungi and fruit flies.

In the Aug. 12 issue of the journal Nature, Caltech computation and neural systems researcher Chris Adami and his colleagues explain the results they have obtained by designing computer programs to be digital organisms that can self-replicate, mutate and adapt by a process analogous to natural selection in nature.

In the study, the authors conclude that complex organisms are more robust than simple ones with respect to how significantly the organisms are affected by single and multiple mutations. Further, the study shows that the overall effect of many mutations can actually result in higher fitness of a complex organism than one would predict from multiplying together the effect of individual mutations.

This latter result tracks closely with experiments with actual simple organisms like bacteria, fungi and fruit flies in which frequent interactions among mutations are observed. But Adami says the conclusions are particularly exciting because the "artificial petri dish" approach demonstrates that digital organisms can be used by researchers to answer important biological questions.

"The advantages are that it's very simple, and that it abstracts the system as much as possible," Adami says. "It's very difficult to ask very fundamental questions about life with a living system because the living system is very complex after four billion years of evolution.

"Life on Earth is all due to one event a long time ago," he says. "Everything we see is related to one accident, so if we look back at this, can we learn something about life in general?"

The answer has implications for future searches for life elsewhere in the solar system and universe, because no one really knows exactly how life got started and how it proceeded to grow in complexity. Therefore, no one really knows all the ground rules of life.

"If we go somewhere else, are we going to find life that is similar or totally different? If it's similar but unrelated, then life is perhaps constrained narrowly. But if it's totally different, then maybe life is constrained very loosely."

Adami's program is based on some of the principles that are known about life and assumed likely to be true elsewhere: living systems replicate, they conserve information and they have dynamic properties that differ from other living systems and allow adaptations.

The digital organisms are based on these principles. By building a digital petri dish in which the programs "live," the researchers can allow the programs to live and fill up a niche, interact with each other, mutate and adapt to local conditions, die out, provide opportunities for other organisms to fill a niche-all the things that organisms on Earth really do, but over many eons.

"We can reconstruct the genetic tree, then change the origin of the tree slightly and rerun the entire tape of evolutionary history," Adami says. "If we change just one molecule way back, this can change everything, we discovered."

The question Adami hopes the new article in Nature will help settle is whether running experiments with digital organisms in a computer is really biology, as biologists understand it. There are skeptics, he says, but he nonetheless thinks the method will gain more believers as the work progresses.

Too, it doesn't hurt that a respected biologist with many years of outstanding accomplishments with actual petri dish cultures is now a collaborator.

"Richard Lenski is the world's expert at doing experimental evolution with E. coli," Adami says. "This paper is the first result of the collaboration, in which we repeated an experiment he has already done with E. coli.

"So I think this is the first time we have convinced biologists that artificial life is not just a pipe dream, but is answering some fundamental questions about biology."

In addition to Adami and Lenski, who is with the Center for Microbial Ecology at Michigan State University, the authors are Charles Ofria, who just earned his doctorate at Caltech in computation and neural systems; and Travis C. Collier of the UCLA Department of Organismic Biology, Ecology and Evolution.