Direct Determination of the Distribution of Fitness Effects of Spontaneous Mutations in Caenorhabditis Elegans

The distribution of fitness effects (DFE) of new mutations is of fundamental importance in numerous areas of theoretical evolutionary biology, as well as having several important practical applications in biomedicine, including complex disease and cancer. Any effort to model population genetic phenomena requires realistic estimates of the relevant parameters, but the fitness effects of (most) mutations in humans cannot be estimated directly, and indirect estimates from standing genetic variation are subject to several sources of ambiguity and bias. Reliable direct estimates of the DFE from tractable model organisms can inform efforts to model human population genetic processes, as well as our understanding of evolution in general. Stocks of the nematode Caenorhabditis elegans that have previously accumulated mutations for 250 generations under relaxed natural selection will be employed to estimate the DFE, combining the methods of classical quantitative genetics with high-throughput genotyping and phenotyping. The twin goals of aim 1 are (1) to obtain a very accurate estimate of the whole-genome mutation rate by deep sequencing a subset of the mutation accumulation (MA) lines with several complementary sequencing technologies, and (2) obtain a very accurate estimate of the cumulative decline in fitness with mutation accumulation by employing large-particle flow cytometry (aka a "worm sorter") to measure lifetime reproduction of a very large number of individual worms. The average decline in fitness divided by the average number of mutations is the average mutational effect on fitness. MA lines will be mated to the unmutated ancestor to generate F1 heterozygotes, which will be similarly characterized for fitness. These experiments will provide the best estimates of the average of the homozygous and heterozygous effects of new, spontaneous mutations in any organism. The goal of Aim 2 is to cross two MA lines and construct a large panel of Advanced Intercross Recombinant Inbred Lines (RIAILs). Deep sequencing of the parental lines will reveal the entire set of mutations carried by each line; each RIAIL can then be genotyped at each potentially mutant locus and individual mutations can be treated as quantitative trait loci (QTL), from which the homozygous DFE can be inferred statistically. Two questions of immediate interest are (1) do mutations of large effect (> a few percent) contribute to the decline in fitness, or is the fitness decline due to a large number of mutations of very small effect, and (2) are beneficial mutations more than trivially rare? The goal of Aim three is to cross each RIAIL constructed in Aim 2 to the unmutated ancestor of the MA lines and estimate fitness in the F1 heterozygotes. The resulting data will provide an estimate of the heterozygous DFE; the heterozygous DFE is the most important for understanding the evolutionary dynamics of rare deleterious mutations. Further, it will be apparent if overdominant mutations (i.e., heterozygote advantage) are more than trivially rare. The extent to which true overdominance is a factor in evolution has been argued for many decades, but unambiguous evidence for true overdominance is almost non-existent. This aim will shed light on that important long-standing question.