Evolutionary bioinformatics: Predicting stability of asexual genomes by global computing

The aim of this work is to study the stability of asexual (non-recombining) genomes. Towards this end several approaches are taken. To investigate the effects of observed, high mutation rates on long-term viability of mitochondria, Mulleris ratchet theory is employed. As analytic approximations allow only predictions for a limited range of the three parameters, effective population size, mutation rate and selection coefficient, individual-based simulations are performed to check and extend the analytic solutions. Results show that Mulleris ratchet might indeed be a threat to mitochondria on a 20 million year timescale and thus appears to complement the threat that is known to come from the high deleterious mutation rates in the nuclear genome. A variety of biological processes promoted to solve this genomic decay paradox are discussed. Reviewing potential solutions shows the enormous need for further simulation models and more computing time to address these issues. Therefore, the design for a software framework is proposed that uses the power of global computing to investigate individual-based models of evolution. The first simulator in a long series of future simulators is implemented and used to start the first global computing system for evolutionary biology, evolution@home (see http://www.evolutionary-research.net). Its results (>28 000 simulations, >16 years CPU time from >200 participants) are used to investigate potential solutions for the discrepancy between high, short-term intergenerational mutation rates observed of human mitochondria and low, long-term mutation rates inferred from phylogenies. Further details of Mulleris ratchet are observed for the first time. The same set of tools that allows quantification of Mulleris ratchet in mtDNA is applied to (non-recombining) microbes, to investigate consequences of deleterious mutations in their genomes. As conclusions critically depend on estimates for deleterious genomic mutation rates, a system is developed for analysis of large numbers of growth curves. This system is used to quantify the deleterious mutation rate of a population of E. coli in a stationary phase mutation accumulation experiment according to the Bateman-Mukai technique. Further observations include unprecedentedly accurate measurements of the maximal growth rate in an overnight culture and potential evolutionary effects of freezing in glycerol at -70ƒC. Detailed analysis shows that Mulleris ratchet might contribute significantly to the majority of uncultivable bacteria.