Berlin 2024 – scientific programme

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

BP: Fachverband Biologische Physik

BP 20: Poster IIIa

BP 20.31: Poster

Wednesday, March 20, 2024, 11:00–14:30, Poster B

Evolution on fitness landscapes with universal negative epistasis — •Daniel Oros and Joachim Krug — University of Cologne, Institute for Biological Physics, Germany

An approach to model evolution is by describing it as a process that selects viable genotypes from the space of all possible ones. Each genotype is assigned a fitness value, corresponding to its reproductive success. The search process then takes place in the landscape defined by the genotype and their corresponding fitness, referred to as fitness landscape. The notion of epistasis refers to the interaction between different mutations in the genotype and their effect on fitness. We investigate certain fitness landscapes that are structured [1]. The concept of universal negative epistasis (UNE) imposes short and long ranged effects on genotype interactions. A way to sample UNE landscapes and an analysis of them is presented together with a connection to the tradeoff induced landscapes (TIL) model [2]. This leads to a lower bound on the number of genotypes from which a peak is accessible from, already known from the TIL-Model. Recent experiments, see [1], highlight the importance of highly accessible landscapes. How their basins of attraction, meaning all genotypes from which a peak is accessible by increasing its fitness by mutations, differ is discussed along with general mathematical properties of UNE.

[1] Krug, J. and Oros, D. (2023) Evolutionary accessibility of random and structured fitness landscapes, https://arxiv.org/abs/2311.17432

[2] Suman G Das et al. Predictable properties of fitness landscapes induced by adaptational tradeoffs. In: eLife 9 (May 2020)

Keywords: Evolution; Statistical Physics; Epistasis; Fitness Landscapes