Title: Polymorphism and the Predictability of Evolution in Fisher’s Geometric Model

12 Predicting the future evolutionary state of a population is a primary goal of evolutionary 13 biology. One can differentiate between forward and backward predictability, where forward 14 predictability is the probability of the same adaptive outcome occurring in independent 15 evolutionary trials, and backward predictability is the likelihood of a particular adaptive 16 path given the knowledge of the starting and final states. Most studies of evolutionary 17 predictability assume that alleles along an adaptive walk fix in succession with individual 18 adaptive mutations occurring in monomorphic populations. However, in nature, adaptation 19 generally occurs within polymorphic populations, and there are a number of mechanisms 20 by which polymorphisms can be stably maintained by natural selection. Here we 21 investigate the predictability of evolution in monomorphic and polymorphic situations by 22 studying adaptive walks in diploid populations using Fisher’s geometric model, which has 23 been previously found to generate balanced polymorphisms through overdominant 24 mutations. We show that overdominant mutations cause a decrease in forward 25 predictability and an increase in backward predictability relative to diploid walks lacking 26 balanced states. We also show that in the presence of balanced polymorphisms, backward 27 predictability analysis can lead to counterintuitive outcomes such as reaching different final 28 adapted population states depending on the order in which mutations are introduced and 29 cases where the true adaptive trajectory appears inviable. As stable polymorphisms can be 30 generated in both haploid and diploid natural populations through a number of 31 mechanisms, we argue that natural populations may contain complex evolutionary histories 32 that may not be easily inferred without historical sampling. 33 3 . CC-BY-NC-ND 4.0 International license peer-reviewed) is the author/funder. It is made available under a The copyright holder for this preprint (which was not . http://dx.doi.org/10.1101/001016 doi: bioRxiv preprint first posted online Dec. 2, 2013;