In what has become a popular line of research, the digital bugs are allowed to reproduce, mutate and compete for computer processing time, in much the same way that real, living organisms compete for nutrients and the energy that gives them.

Changes that might take millions of years to observe in the real world can now be seen on a more useful time scale, according to the program's creators, Richard Lenski and colleagues from the Michigan State University in the US.

Another advantage over similar experiments involving fast-breeding organisms like bacteria is that each minute step of evolution in each digital bug can be recorded, so that a complete evolutionary history is available.

Primordial soup

"This artificial world yields some of the same complexities we see in the real living world, but we have trouble studying these complexities in detail with the real organisms because the genetic experiments get too complicated," Lenski said.

Inman Harvey of the University of Sussex in southern England has reviewed the research, which appears in the journal Nature, and said the artificial world is like a primordial soup.

"You've got these digital organisms evolving away and you can just pick one out and measure how well it replicates. You can then take a clone and put a mutation in."

Lenski said the marriage of computer science and biology allowed the researchers to study what happened when organisms were exposed to more than one genetic mutation.

Additive effects

"One of the questions, we wanted to ask...is whether the effects of mutations can be understood in isolation or whether, if you have two mutations, the combined effect might be different from the sum of the two," Lenski said.

His team found that multiple 'mutations' occurring at random in the digital organisms did not necessarily have simple additive effects on the organisms' ability to multiply and carry out simple 'functions'.

Instead, as in real-life organisms, the different mutations interacted with each other to affect performance in more unexpected ways.

They also found that the more complex digital bugs tended to show a greater resistance to multiple mutations.

Different effects

"Simple bugs just got sicker and sicker. For the more complex ones, the rate at which they got sicker began to slow down. The effects of combining two mutations did have a different effect," Lenski said.

"Our results suggest that there might be some relation between sensitivity to multiple mutations and genome size."

Some scientists remain to be convinced of the value of such virtual worlds, doubting whether digital bugs are sufficiently sophisticated to tell us anything useful about real-world evolution.

"If you can trust it to be valid, then you are laughing. However that is a big if," said Inman Harvey. "Without a doubt, digital organisms mirror living organisms in particular ways but the question is how far you can push that.