Dresden 2026 – wissenschaftliches Programm
Bereiche | Tage | Auswahl | Suche | Aktualisierungen | Downloads | Hilfe
BP: Fachverband Biologische Physik
BP 21: Focus Session: Sequence Spaces, Populations and Evolution
BP 21.6: Vortrag
Mittwoch, 11. März 2026, 17:00–17:15, HÜL/S386
A simplified Rough Mount Fuji model clarifies how local adaptive walks can reach the highest peaks in rugged fitness landscapes — •Kye E Hunter1,2 and Nora Martin1 — 1CRG (Barcelona Collaboratorium for Modelling and Predictive Biology) — 2Facultat de Física, Universitat de Barcelona (UB)
Adaptive evolution selects random genotypic mutations according to their fitness. This can be modeled using a fitness landscape, a network of possible genotypes with a fitness value associated to each sequence. In the simplest models of adaptive evolution, populations move through this network in fitness-increasing steps until reaching a genotype whose fitness exceeds that of all its neighbors—a fitness peak. In evolutionary simulations on their empirical folA landscape, Papkou et al. (Science 2023) found that such fitness-increasing walks are likely to reach the globally-highest-ranked peaks among a large number of peaks, despite only being based on local rules. Similar results were found in a mathematical model of fitness landscapes, the Rough Mount Fuji model (Li & Zhang MBE 2025).
We use a simplified Rough Mount Fuji model to find simple analytical explanations for how a landscape can have both a large number of peaks and populations that reach the highest-ranked peaks. Our explanation proceeds by dividing the landscape into different regions, and considering the number of peaks relative to the total number of genotypes in each region. We then identify the degree to which those same arguments apply in the empirical folA landscape.
Keywords: Evolution; Fitness landscape; Rough Mount Fuji; Deep mutational scan