Part 1: Adaptation in a changing environment from de-novo mutations.
Project RACE began with the translation of the methods commonly used by computer scientists into biological terminology. EC is a sub-field of computer science that uses evolutionary processes for optimization and design. Many powerful techniques have been developed to analyze evolutionary algorithms, with strengths and weaknesses complementary to methods commonly used in evolutionary biology (EB).
Recently established collaboration between both fields allowed identifications of many equivalents between them: while EB deals with DNA, genes, loci, and alleles, direct equivalents in EC exist - bitstrings, bits, zeros, and ones. A proper translation of the problems enables the use of many methods developed and widely used in EC to analyze EB problems.
I created a complex model of a population of organisms, adapting to an environment that was changing at various frequencies. Unlike in many previous studies, the organisms had multiple genes that contributed to their fitness and, thus, survival. Different modes of environmental changes were also considered. In collaboration with computer scientists M. Krejca, T. Kötzing, and P.K. Lehre, we investigated adaptation rates and limits, as well as effects of natural parameters, such as population sizes and the mode of environmental. We showed that while frequent change between environmental conditions hinders adaptation, it may, under certain conditions, protect the population from extinction. This happens when frequent environmental change prevents the population from overspecialization, resulting in populations of generalists that may survive, though not thrive, in a wide range of conditions.
Published in: Trubenová, B., Krejca, M.S. Lehre, P.K. and Kötzing, T. (2019): Surfing on the seascape: Adaptation in a changing environment. Evolution, 73: 1356-1374. doi:10.1111/evo.13784
Part 2: Evolution in changing social environment
For many organisms, part of the environment is social: provided by family and group members, or even different species. Social interactions are universal in nature and have profound effects on evolutionary processes. For instance, predator-prey interactions are also considered social and are well known to influence both participating species' evolution. The social environment itself is both the actor and the subject to selections and may thus change, resulting in a feedback loop.
I focused on a well-known scenario of fictitious green-beards and the evolution of altruism. In the green-beard concept, a single gene or several tightly linked genes encoding altruistic behavior must meet three requirements:
1. Cause its bearer to behave altruistically.
2. Display an observable and distinctive trait (the 'green beard').
3. Be able to distinguish between individuals that do not display the trait and those that do.
Such a gene can recognize individuals with copies of itself, helps them, and so helps to propagate itself. As the green-beard associated altruistic behavior evolves, the environment that individuals experience changes. Whether these changes will lead to the extinction or survival of the altruistic individuals is an interesting question, puzzling biologists for decades.
I used analytical models and simulations of social interaction to determine the conditions under which the altruistic individuals persist and thrive, and what changes this will mean for the social environment. I derived the minimum correlation between the signaling trait (a green beard) and the altruistic trait required for the latter's evolution and determined its dependence on various parameters, such as the strength of the interaction, costs, and benefits of the altruistic behavior. I further showed the conditions and probability of the extinctions of altruistic populations caused by the invasion by selfish individuals.
Published in: Trubenová, B, Hager, R.(2019): Green beards in the light of indirect genetic effects. Ecol Evol. 2019; 9: 9597– 9608.
https://doi.org/10.1002/ece3.5484(opens in new window)Part 3: Predicting current genetic diversity from historical data
Changing environment means changes in the selection pressure: its direction and strength. This leads to changes in population sizes, even extinction. Understanding the evolutionary and ecological factors that determine populations' ability to adapt to a changing environment is critical for conservation efforts. Adaptation is possible either by de-novo mutations (investigated in the first part of the project) or from the standing genetic variation in the population. Therefore, studying genetic diversity and understanding its dependence on environmental changes is essential.
In the third part of the project, in collaboration with Katalin Csillery and Eniko Szep, I focused on the effect of environmentally determined demographic changes on the genetic diversity of populations. We developed a spatially explicit coalescent simulation tool called gridCoal, based on Python msprime package. In gridCoal, environmental changes are represented by defined demographic histories (population sizes defined in time and space) and migration patterns and other important population parameters. The simulator uses coalescence simulations to estimate the genetic diversity, approximated from coalescence times, across a defined grid.
We used gridCoal to investigate various scenarios of spatial and temporal changes in population sizes and compared the results with theoretical predictions and experimental data whenever possible. Simulations allowed us to examine the influences of such scenarios as a population undergoing a bottleneck caused by Ice Age, or expansions following up warmer and more favorable conditions.
The gridCoal simulator can be used to predict areas, or zones, with increased genetic diversity. Therefore, we hope that our simulator will be used by biologists working in the field of conservation to make more informed decisions needed for species protection and nature conservation.
