But there are well-adapted, complex organisms, orchids, humans, you name it, and confusion about 'cost of complexity' offers ammunition to proponents of Creationism, who hold that intricacy could only arise only through the efforts of a divine designer and not natural selection.
A new analysis says it reveals flaws in the models from which the cost of complexity idea arose and shows that complexity can develop through evolutionary processes. They instead say a moderate amount of complexity best equips organisms to adapt to environmental change,
The study focused on a genetic phenomenon called pleiotropy, in which a single gene affects more than one trait. Examples of pleiotropy are well known in certain human diseases, and the effect also has been documented in experimental animals such as fruit flies. Biologists also recognize its importance in development, aging and many evolutionary processes. However, pleiotropy is difficult to measure, and its general patterns not easily understood.
Scientists have developed mathematical models of the phenomenon, based on certain assumptions, and have made predictions from the results of the models. A group of researchers decided to test the assumptions against real-life observations by analyzing several large databases that catalog the effects of specific genetic mutations on traits in model organisms (yeast, roundworms and mice). Each data set included hundreds to thousands of genes and tens to hundreds of traits.
For simplicity, mathematical models of pleiotropy have assumed that all genes in an organism affect all of its traits to some extent. But they found found that most genes affect only a small number of traits, while relatively few genes affect large numbers of traits.
What's more, they found a "modular" pattern of organization, with genes and traits grouped into sets. Genes in a particular set affect a particular group of traits, but not traits in other groups.
In addition, the researchers learned that the more traits a gene affects, the stronger its effect on each trait.
All of these findings challenge the assumptions underlying the classic mathematical models that suggest complexity is prohibitively costly. When Fisher first wrote about the cost of complexity, he argued that random mutations, which, along with natural selection, drive evolution, are more likely to benefit simple organisms than complex organisms.
"Think of a hammer and a microscope," says Jianzhi Zhang, co-author of the study from the University of Michigan "One is complex, one is simple. If you change the length of an arbitrary component of the system by an inch, for example, you're more likely to break the microscope than the hammer."
In a paper published in 2000, evolutionary geneticist H. Allen Orr of Rochester came up with additional reasons for the cost of complexity. According to his model, even if a mutation benefits a complex organism, it's unlikely to spread throughout the whole population and become "fixed." And even if it does that, the advantage of the mutation is likely to be small.
By incorporating a more realistic representation of pleiotropy, the new analysis found the reverse of Orr's arguments to be true. Although Fisher's observation still holds, reversing Orr's assertions minimizes its impact, thus reducing the cost of complexity.
Further, the analysis showed that the ability of organisms to adapt is highest at intermediate levels of complexity. "This means a simple organism is not best, and a very complex organism is not best; some intermediate level of complexity is best in terms of the adaptation rate," Zhang said.
The new findings help buffer evolutionary biology against the criticisms of intelligent design proponents, Zhang said. "The evolution of complexity is one thing that they often target. Admittedly, there were some theoretical difficulties in explaining the evolution of complexity because of the notion of the cost of complexity, but with our findings these difficulties are now removed."
Citation: Zhi Wang, Ben-Yang Liao, Jianzhi Zhang, 'Genomic patterns of pleiotropy and the evolution of complexity', PNAS published ahead of print September 27, 2010, doi:10.1073/pnas.1004666107