Genetically integrated traits and rugged adaptive landscapes in digital organisms

摘要: When overlapping sets of genes encode multiple traits, those traits may not be able to evolve independently, resulting in constraints on adaptation. We examined the evolution genetically integrated digital organisms—self-replicating computer programs that mutate, compete, adapt, and a virtual world. assessed whether overlap encoding two – here, ability perform different logic functions constrained also strong opposing selection could separate otherwise entangled allowing them independently optimized. Correlated responses were often asymmetric. That is, increase one function produced correlated response other function, while second caused complete loss first function. Nevertheless, most pairs successfully disentangled when was applied break apart. In an interesting exception this pattern, AND evolved counter its optimum some populations owing EQU Moreover, showed strongest only after it from AND, such lost. Subsequent analyses indicated against had altered local adaptive landscape cross what would have been valley thereby reach higher fitness peak. can sometimes constrain However, our study, even strongly usually insufficient impose long-lasting constraints, given input new mutations fueled selective responses. detailed information about useful for predicting outcome traits. Finally, results illustrate richness evolutionary dynamics systems highlight their utility studying processes thought important biological systems, but which are difficult investigate systems.