Beneficial Fitness Effects Are Not Exponential for Two Viruses

• DOI: 10.1007/s00239-008-9153-x
• Publication Date: 2008-09-08T20:54:40Z
• Authors: Darin R. Rokyta; Craig J. Beisel; Paul Joyce; Martin T. Ferris; Christina L. Burch; Holly A. Wichman
• Keywords: Gumbel distribution

The distribution of fitness effects for beneficial mutations is paramount importance in determining the outcome adaptation. It generally assumed that follow an exponential distribution, example, theoretical treatments quantitative genetics, clonal interference, experimental evolution, and adaptation DNA sequences. This assumption has been justified by statistical theory extreme values, because fitnesses conferred should represent samples from right tail distribution. Yet value theory, there are three different limiting forms tails distributions, describes only those distributions Gumbel domain attraction. Using two viruses, we show first time can be rejected favor a with right-truncated tail, thus providing evidence upper bound on effects. Our data also violate common small-effect greatly outnumber large effect, as they consistent uniform