Evolution at Molecular Resolution Simulation of In Vitro RNA Selection

A new concept of evolution extends conventional population genetics in two aspects: (i) the notion of sequence space which is required to describe migration of populations through adaptive walks and random drift, and (ii) genotype - phenotype relations expressed in the case of test-tube evolution by the mapping of RNA sequences into structures. This concept, so far, has been developed for molecular evolution of RNA. RNA structures are commonly simplified as secondary structures. First attempts to describe the course of evolutionary optimization of tRNA structures from random sequences in full molecular detail were successful. These previous studies will be supplemented and extended by (i) the statistical analysis of computer simulation experiments on RNA opimization, (ii) the development of an analytical stochastic model of the evolutionary process, (iii) the development of a "statistical topology" that allows to define nearness and attainability of phenotypes within an evolutionary context, and (iv) extensions of the current model to the design of RNA molecules with predefined properties, three-dimensional structures of RNA as well as proteins. Particular attention will be given to applications of the current concept of evolutionary dynamics to the design of biomolecules with new properties. Examples are the development of efficient protocolls in evolutionary biotechnology together with experimental groups.

The details of evolutionary optimization of RNA structures in test-tube experiments were investigated by means of computer simulation and mathematical analysis. The main result of these studies is an approximately constant number of main transitions in the course of the evolutionary process, which are accompanied by large structural changes. The number of main transitions or structural innovations depends only on the initial and the final state of the process. The new conformations are contained at high frequency in the mutation spectrum of the transition sequence, whereas they are rare otherwise. A major progress in rational design of RNA molecules is the development of an algorithm for the search of molecular switches. These are sequences that switch between two conformations with different properties. An experimental example for such a switch (Schultes & Bartel, Science 289:448, 2000) consists of a designed RNA molecule, which exhibits two different catalytic functions.