Computer programs ‘mutate’ to outlast viruses

"Even though we’re basically studying computer viruses, we're able to address some of the oldest questions in evolutionary biology," Luis Zaman says. (Credit: Bruno Cordioli/Flickr)

The KISS concept—keep it simple, stupid—may work in many situations. However, when it comes to evolution, complexity appears to be the best strategy for longterm survival.

Researchers pitted self-replicating computer programs against computer viruses in the domain of the Avida platform for digital evolution.

The co-evolution of host and virus ultimately led to organisms with capabilities superior to those of organisms that evolved without battling the viruses, says Luis Zaman of Michigan State University’s BEACON Center for the Study of Evolution in Action.

“The organisms faced off against some pretty nasty viruses, ones that quickly overcame the easy ways of becoming resistant,” says Zaman, who’s now a postdoctoral researcher at the University of Washington.

“This left only more and more complex options for co-evolving hosts. We thought to ourselves, maybe the organisms that were more readily adaptable than the others, more evolvable, would be the ones left standing at the end of the experiments.”

Key mutations

The team of scientists showed that the long-lasting hosts were indeed more evolvable. The results are detailed in a study published in PLOS Biology.


They measured how frequently random mutations in the hosts’ genetic code provided escape from infection. The scientists found a surprisingly large proportion of these so-called “switching” mutations in the co-evolved hosts—nearly 10 times more than their counterparts that hadn’t co-evolved.

They attained these key mutations by deftly dancing with the very agent trying to kill them. Parasite attacks, host evolves, counters, and survives. Parasite encodes and thereby metaphorically reviews the entire history of the war in its genetic code, advances and deploys a new weapon. Host blocks with shield. And on and on the fighting rages.

The war doesn’t deplete the hosts. The battle-tested organisms instead survive, but only the strong are left. Rather than exile the enemy, hosting the virus is perhaps a better strategy in the long run—one with ramifications well beyond the digital model.

These computer parasites provide a model that may explain how evolutionary arms races drive the emergence of complexity.

“Even though we’re basically studying computer viruses, we’re able to address some of the oldest questions in evolutionary biology,” Zaman says. “Co-evolution and parasites are everywhere, and our research with Avida lets us understand their effects on complexity and evolvability with incredible detail.”

The National Science Foundation funded the research. Zaman conducted the study with fellow MSU BEACON scientists Justin Meyer, now at the University of California-San Diego; Suhas Devangam, who’s now at Wayne State University; David Bryson, who’s now at Apple; Richard Lenski and Charles Ofria.

Source: Michigan State University