Towards evolutionary predictions: current promises and challenges

Abstract

Evolution has traditionally been a historical field of study and predicting evolution has long been considered challenging or even impossible. However, evolutionary predictions are increasingly being made and used in many situations in medicine, agriculture, biotechnology and conservation biology. Because every field uses their own language and makes predictions from their background, researchers are not always aware of the breadth of evolutionary predictions. Evolutionary predictions may be used for several purposes such as to prepare for the future, to try and change the course of evolution or simply to determine how well we understand an evolutionary system. Exactly what aspect of an evolving population we want to predict, such as the most common genotype, average or individual fitness, or population size, depends on the situation. There are many uses of evolutionary predictions that may not be recognized as such. Therefore, the main goal of this review is to increase awareness of methods and data that are used to make these predictions in different fields, by showing the breadth of situations in which evolutionary predictions are made. We describe how evolutionary predictions are highly diverse, but nevertheless share a common structure described by the predictive scope, horizon, precision and risk. Then, by using examples ranging from SARS-CoV2 and influenza to CRISPR-based gene drives and sustainable product formation by microorganisms, we discuss the methods for predicting evolution, factors that affect the predictability, and how predictions can be used to prevent unwanted evolution or promote beneficial evolution. We hope that this review will increase collaboration between fields by creating a common language for evolutionary predictions.

Epistasis and Adaptation on Fitness Landscapes

“…Recent advances in technology and experimental approaches have enabled the empirical assessment of increasingly large empirical fitness landscapes and the subsequent evaluation of fitness landscape theory. In scenarios in which selection is strong (e.g., in large microbial populations), and if these findings translate to natural environments, empirically tuned fitness landscape models promise to be useful for predicting evolution, which is relevant in human medicine (e.g., antimicrobial resistance or cancer evolution) and with respect to climate change (e.g., potential for evolutionary rescue under the climate crisis) (de Visser & , Lässig et al 2017, Wortel et al 2021.…”

The extinction time under mutational meltdown

“…The extinction of a population is a fundamental process in evolutionary biology and, given its irreversible nature, a lot of work across scientific fields has been devoted to its prediction (Ovaskainen and Meerson 2010; Carlson et al 2014; Matuszewski et al 2017; Wortel et al 2021). For example, in medicine, the extinction time of a pathogen can decide whether its host survives or dies, whereas in conservation biology, dooming extinction calls for immediate action.…”

The extinction time under mutational meltdown driven by high mutation rates

“…The extinction of a population is a fundamental process in evolutionary biology and, given its irreversible nature, a lot of work across scientific fields has been devoted to its prediction (Carlson et al, 2014; Matuszewski et al, 2017; Ovaskainen & Meerson, 2010; Wortel et al, 2021). For example, in medicine, the extinction time of a pathogen can decide whether its host survives or dies, whereas in conservation biology, looming extinction calls for immediate action.…”