A testable genotype-phenotype map: modeling evolution of RNA molecules

Abstract

Recent experiments and progress in modelling evolution in silico converge towards a coherent view of Darwinian evolution in molecular systems. Conventional population genetics and quasi-species theory model evolution in genotype space and properties of phenotypes enter evolutionary dynamics as parameters only. RNA evolution in vitro is an appropriate basis for the development of a new and comprehensive model of evolution, which is focussed on the phenotype and its fitness relevant properties. Relation between genotypes and phenotypes are described by mappings from genotype space onto a space of phenotypes. These mappings are many-to-one and thus give ample room for neutrality. The RNA model reduces genotype-phenotype relations to a mapping from sequences into secondary structures with minimal free energies and allows to derive otherwise inaccessible quantitative results. RNA sequences that fold into the same structure form neutral networks in genotype space, which determine the course of evolution. Neutral networks are embedded in sets of compatible sequences. Intersections of these sets represent regions in sequence space where single molecules can form two or more structures. Continuity and discontinuity in evolution are defined through straightforward interpretation of computer simulations of RNA optimization. In silico evolution provides insight into the accessibility of phenotypes and demonstrate the constructive role of random genetic drift in the search for phenotypes of higher fitness. New experimental data, among them the results of genome research, will present a solid basis for test and further development of the model for phenotype evolution.