Parts | Days | Selection | Search | Updates | Downloads | Help

DY: Fachverband Dynamik und Statistische Physik

DY 32: Statistical Physics in Biological Systems III (organised by BP)

DY 32.5: Talk

Thursday, March 17, 2011, 11:45–12:00, ZEU 260

Assessing the asymptotic fitness distribution of beneficial mutations from incomplete data sets — •Ivan G. Szendro¹, Martijn Schenk², J. Arjan G.M. de Visser², and Joachim Krug¹ — ¹Institut für Theoretische Physik, Universität zu Köln — ²Laboratory of Genetics, Wageningen University

Since seminal work by Gillespie [1] and Orr [2] it is expected that the distributions of fitness effects of beneficial mutations are determined by the universality classes of extreme value theory. More specifically, it is commonly assumed that the distributions of fitness fall into the Gumbel class, implying an exponential decay at large values. However, there have been recent claims that for some viruses the distribution belongs to the Weibull class [3].

In this contribution, we assess the effect of not observing existing beneficial mutations on the assignment of fitness distributions to one of the three extreme value classes. We assume that the probability to observe a specific mutant depends on its selective disadvantage with respect to the fittest observed mutants. In the light of our considerations, we analyze data collected in an experimental evolution study of the TEM-1 β-lactamase enzyme, which confers antibiotic resistance to Escherichia coli.

[1] J.H. Gillespie, Theor. Popul. Biol. 23, 202 [2] H.A. Orr, Genetics 163, 1519 [3] D.R. Rokyta et al., J. Mol. Evol. 67, 368